AI Chat Free Online

AI Chat Free Online — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  YNAB

  You Need a Budget (YNAB) (pronounced ) is an online personal budgeting program based on the envelope system developed by a privately owned American company of the same name. It is available via any web browser or a mobile app. == History == The program was initially developed as standalone software in 2004 by Jesse Mecham, while he was in college pursuing his master's degree in accounting, after he and his wife faced financial difficulty and decided to improve their budgeting. It evolved from a spreadsheet that he created for the budgeting process. The acronym stands for "you need a budget." In 2015 they changed their licensing model to software as a service. In 2020, YNAB had 115 employees, all working remotely. == Overview == The service encourages users to follow four principles or "rules": Give every dollar a job: Each dollar in a budget is allocated to a specific purpose. This concept is also called zero-based budgeting. Embrace true expenses: All expenses are planned for, so that there are no surprises. Roll with the punches: Being flexible when there is overspending. Age your money: Keeping money in your budget without immediately spending it. Users can either import transactions automatically from their financial institutions or input them manually. The software also displays financial reports to keep users informed about their finances at a glance. == Awards and recognition == YNAB has been named one of the best budgeting apps by U.S. News & World Report, Kiplinger's Personal Finance, CNN, HuffPost, CNBC, and hundreds of other financial reporting outlets. The Wall Street Journal – Best budgeting app for hands-on budgeters. Forbes – Best Budgeting Apps Money – Best budgeting app for college students. Lifehacker – Most popular personal finance software. Wirecutter – "Great pick for hard-core budgeters". Investopedia – Best overall budgeting app.

  Read more →
• 

  Data exhaust

  Data exhaust (also exhaust data) is the trail of data generated as a by-product of users' online activity, behaviour, and transactions, rather than data they deliberately create or submit. It forms part of a broader category of unconventional data that also includes geospatial, network, and time-series data, and may be useful for predictive analytics. Data exhaust can take the form of cookies, temporary files, log files, clickstream records and stored preferences. Actions such as visiting a web page, following a link, or dwelling on an element may all generate exhaust data that is recorded without the user's active awareness. Unlike primary content — which the user intentionally creates — exhaust data is a passive side effect of interaction. A bank, for example, might treat the amounts and parties involved in a transaction as primary data, while secondary data could include whether the transaction was carried out at a cash machine rather than a branch. == Uses == Data exhaust collected by companies is often information that is not immediately useful in isolation, but can be aggregated and analysed to improve products, personalise content, identify trends, and support quality control. Companies may also store exhaust data for future analysis or sell it to third parties. Shoshana Zuboff has described this practice as a core mechanism of what she terms surveillance capitalism, in which behavioural data generated by users is converted into predictive products. Kosciejew notes that large quantities of often raw data are collected in this way, much of which is never analysed. == Medical exhaust data == Many medical devices — including pacemakers, dialysis machines and surgical cameras — generate exhaust data as a by-product of their operation. The majority of this data is never captured or analysed, and is typically discarded once a procedure ends or a device completes its routine monitoring cycle. The potential use of data generated by implanted devices such as pacemakers raises additional legal and ethical questions around ownership and consent. Using electronic health records for research also creates challenges because of the volume of data involved, creating a need for automated algorithms to process it. == Privacy and regulation == The collection and distribution of data exhaust is not in itself illegal in most jurisdictions, but its use raises questions of privacy and informed consent. Steps commonly taken to address these concerns include data anonymisation, offering users an opt-out from the sale of their data, and publishing explicit privacy policies that disclose what data is collected and how it is used.

  Read more →
• 

  Caste census

  Caste census is a proposed census to be conducted in India by the Central Government of India. The proposed census was decided under the leadership of Prime Minister Narendra Modi by the cabinet committee of political affairs (CCPA) on 30 April 2025. It has been decided that a caste enumeration should be included with the forthcoming census. The exact time has not been declared yet. It is unclear that when the next census will be held. The decision of the cabinet was announced by the Central Railway Minister Ashwini Vaishnaw. It has been seen as a step that would help in drafting "equitable and targeted" policies by the present Central Government of India led by the Bhartiya Janta Party in India. The Central Home Minister Amit Shah has described the decision as a "historic decision". He has also described that the historic decision as “committed to social justice”. The leader of opposition Rahul Gandhi has welcomed the decision. He said "We have shown we can pressure govt" He has demanded a clear timeline for its completion. He has called it "The first step towards deep social reform". == Description == The caste census is a systematic recording of individuals’ caste identities during the nationwide census in the country. The Central minister Ashwini Vaishnaw expressed his view on the proposed census and said that it would "strengthen the social and economic structure of our society while the nation continues to progress”. The Caste census will happen for the first time in 100 years by the Central Government of India. It will be the part of the upcoming census in India. == History == According to Peabody, the first systematic caste-wise enumeration of households in the Indian subcontinent was conducted between 1658 and 1664 across seven districts of the then Marwar Kingdom, including Jodhpur city which was its capital. It was conducted by the then home minister Munhata Nainsi of the kingdom for the purpose of tax documentation. It was not to for classification of society or creation of social hierarchies but solving a tax related problem. During the period of the British rule in India, caste census was included in the decadal censuses to categorise the population by caste, religion and occupation. In 1871–72, the first detailed caste census was conducted by the government of British Raj in India. It was practiced between the period 1881 to 1931. The last caste census was conducted in the year 1931 in which 4,147 castes were recorded. The largest population in the whole of British India (including Pakistan and Bangladesh) was of Brahmins. The population of Brahmins was recorded more than 1.5 crores. After Brahmin community, the second place was of Jatav (Chamar)community. The population of Jatav was a little more than 1.23 crores. On the third place were Rajputs. The population of Rajputs was 81 lakhs. The Rajput caste was followed by the Kunbi caste of Maharashtra. The population of Kunbi caste was 64 lakhs and 34 thousands. The Kunbi caste was followed by Yadav (Ahir) caste. The population of Yadav (Ahir) community was 56 lakhs and 82 thousands. The Yadav (Ahir) caste was followed by Teli community. The population of Teli community was 42 lakhs and 58 thousands. The Teli community was followed by Gwala community. The population of the Gwala community was 40 lakhs. After the independence of India, the caste enumeration was stopped by the newly independent Government of India led by the prime minister Pandit Jawahar Lal Nehru in 1951. The caste enumeration was stopped to avoid reinforcing social divisions in the Indian society. But, there was an exception made for the enumeration of the Scheduled Castes (SCs) and Scheduled Tribes (STs) in the decadal censuses. Therefore, the enumeration of the Scheduled Castes and the Scheduled Tribes is being conducted in every census since 1951. In 1961, the Government of India permitted states for conducting their own surveys to compile OBC lists, but national caste census was not conducted.

  Read more →
• 

  Thirst trap

  A thirst trap is a type of social media post intended to entice viewers sexually. It refers to a viewer's "thirst", a colloquialism likening sexual frustration to dehydration, implying desperation, with the afflicted individual being described as "thirsty". The phrase entered into the lexicon in the late 1990s, but is most related to Internet slang that developed in the early 2010s. Its meaning has changed over time, previously referring to a graceless need for approval, affection or attention. == History == The term thirst trap originated within selfie culture, though its precise origins remain unclear. An early use of the phrase with reference to dehydration appears in the 1999 book Running for Dummies by Florence Griffith Joyner and John Hanc, where it referred to the deceptive sensation of thirst being quenched after initial fluid intake, advising continued hydration to avoid the so-called "thirst trap." The modern usage of thirst trap resurfaced around 2011 on platforms such as Twitter and Urban Dictionary, coinciding with the growing popularity of Snapchat, Instagram, and dating apps like Tinder and Grindr. In 2011, Urban Dictionary defined it as "any statement used to intentionally create attention or 'thirst'." By 2018, the term had entered mainstream discourse, appearing in outlets such as The New York Times and GQ without the need for explanation. == Usage of the term == Often, the term thirst trap describes an attractive picture of an individual that they post online. Thirst trap can also describe a digital heartthrob. For instance, former Canadian prime minister Justin Trudeau has been described as a political thirst trap. It has also been described as a modern form of "fishing for compliments". == Motivation == Thirst trapping may be driven by a variety of motives. Individuals often seek attention through "likes" and comments on social media, which can offer a temporary sense of validation and improved self-esteem. It can also serve as an outlet for expressing one's sexuality or enhancing a personal brand. In some cases, sharing such content may provide financial gain. Others might post thirst traps to cope with emotional distress, such as after breakup, or to spite a former lover. Sharing a thirst trap has also been used as a way to connect in times of social isolation (e.g. COVID-19 pandemic). From a physiological standpoint, endorphins and neurotransmitters like oxytocin and dopamine are released during sexual contact. It has been speculated outside of the academic setting that sharing and engaging with thirst traps may elicit similar pleasure responses. == Methodology == Methodologies have developed to take an optimal thirst trap photo. Reporting for Vice magazine, Graham Isador found several of his social network contacts spent a lot of time considering how to take the best photo and what text they should use. They considered angles and lighting. Sometimes they made use of the self-timer feature available on some cameras. Often, body parts are put on display without being too explicit (e.g. bulges of male genitalia, breast cleavage, abdominal muscles, pectoral muscles, backs, buttocks). Often, the thirst trap is accompanied by a caption. For instance, in October 2019, actress Tracee Ellis Ross posted bikini pictures on Instagram with a caption that included the message: "I've worked so hard to feel good in my skin and to build a life that truly matches me and I'm in it and it feels good. ... No filter, no retouch 47 year old thirst trap! Boom!" On Instagram, #ThirstTrapThursdays is a popular tag. Followers reply in turn after a posting. == Variations == "Gatsbying" is a variation of the thirst trap, where one puts posts on social media to attract the attention of a particular individual. The term alludes to the novel The Great Gatsby where the character Jay Gatsby would throw extravagant parties to attract the attention of his love interest, Daisy. "Instagrandstanding" is an alternative name for this. "Wholesome trapping" has developed, where one posts pictures of more meaningful aspects of life, such as spending time with friends or doing outdoor activities. == Criticism == Psychotherapist Lisa Brateman has criticized thirst traps as an unhealthy method of receiving external validation. This desire for external validation can be addictive. Thirst traps can cause pressure to maintain a good physical appearance, and therefore cause self-esteem issues. Additionally, thirst traps are often highly choreographed and thus present a distorted perception of reality. The manufacturing of thirst traps can be limited when one enters a relationship or with time as the body ages. In some cases, thirst traps can lead to harassment and online bullying. In April 2020, model Chrissy Teigen posted a video of herself wearing a black one-piece swimsuit, and she received a multitude of negative comments that constituted bullying and body shaming.

  Read more →
• 

  Artificial intelligence in education

  Artificial intelligence in education (often abbreviated as AIEd) is a subfield of educational technology that studies how to use artificial intelligence to create learning environments. Considerations in the field include data-driven decision-making, AI ethics, data privacy and AI literacy. Concerns include the potential for cheating, over-reliance, equity of access, reduced critical thinking, and the perpetuation of misinformation and bias. == History == Efforts to integrate AI into educational contexts have often followed technological advancement in the history of artificial intelligence. In the 1960s, educators and researchers began developing computer-based instruction systems, such as PLATO, developed by the University of Illinois. In the 1970s and 1980s, intelligent tutoring systems (ITS) were being adapted for classroom instruction. The International Artificial Intelligence in Education Society was founded in 1993. Coinciding with the AI boom of the 2020s, the use of large language models in the global north has been promoted and funded by venture capital and big tech. Companies creating AI services have targeted students and educational institutions as customers. Similarly, pre-AI boom educational companies have expanded their use of AI technologies. These commercial incentives for AIEd use may be related to a potential AI bubble. In the U.S., bipartisan support of AI development in K-12 education has been expressed, but specific implementations and best practices remain contentious. == Theory == AIEd applies theory from education studies, machine learning, and related fields. A 2019 review of the previous decade of studies found that most research prioritized technological design over pedagogical integration. Ouyang and Jiao (2021) propose three paradigms for AI in education, which follow roughly from least to most learner-centered and from requiring least to most technical complexity from the AI systems: AI-directed, learner-as-recipient: AIEd systems present a pre-set curriculum based on statistical patterns that do not adjust to learner's feedback. AI-supported, learner-as-collaborator: Systems that incorporate responsiveness to learner's feedback through, for example, natural language processing, wherein AI can support knowledge construction. AI-empowered, learner-as-leader: This model seeks to position AI as a supplement to human intelligence wherein learners take agency and AI provides consistent and actionable feedback. Some scholars place AI in education within a socio-technical framework. This positions AI alongside other emerging educational technologies, such as computing, the internet, and social media. The framework of Tsao, Heinrichs and Camit (2025) draws on new materialism and posthumanism, specifically Donna Haraway's concept of sympoiesis (making-with). This perspective views learning as an entanglement of human and non-human actors (students, teachers, and AI algorithms), where knowledge is co-composed in contact zones between human context and algorithmic prediction. AI agents have been trained on biased datasets, and thus continue to perpetuate societal biases. Since LLMs were created to produce human-like text, algorithmic bias can be introduced and reproduced. AI's data processing and monitoring reinforce neoliberal approaches to education rather than addressing inequalities. == Applications == Uses of generative AI chatbots in education have included assessment and feedback, machine translations, proof-reading exam question generation and copy editing, or as virtual assistants. Emotional AI in education is the study and development of systems that can detect learners' emotions or provide emotional support in learning. == Usage == === Schools and educators === Following the release of ChatGPT in November 2022, some schools and large school districts blocked access to the site and issued warnings that the use of such tools would be seen as cheating. Governmental and non-governmental organizations such as UNESCO, Article 4 of the European Union's AI Act, and the U.S. Department of Education have published reports advocating for specific AIEd approaches. National higher-education bodies have also published guidance on generative AI, including Ireland's Higher Education Authority, which issued a policy framework for higher education teaching and learning in December 2025. In 2024, UNESCO released updated global guidance for generative AI in education, emphasizing ethical use, teacher training, and data protection to ensure responsible integration of AI tools in learning environments. According to Taso (2025), policy implementation in higher education is interpreted and enacted differently by various organizations. These decentralized policies can lead to inconsistent enforcement and confusion among students regarding what constitutes acceptable use, with the burden of ethical navigation falling on individual teachers and students. AI integration in classrooms has created new forms of invisible labour for educators, who must navigate ambiguous policies, redesign assessments to be AI-resilient, and adjudicate potential academic integrity violations. The use of AI detection tools has also been criticised for creating an adversarial relationship between students and institutions, where students may be falsely accused of misconduct based on probabilistic software. AIEd advocates say that efforts should be made towards increasing global accessibility and training educators to serve underprivileged areas. === Students === Reliance on generative AI has been linked with reduced academic self-esteem and performance, and heightened learned helplessness. Algorithm errors and hallucinations are common flaws in AI agents, making them less trustworthy and reliable. According to a 2025 survey from Inside Higher Ed, 85% of higher education students use generative AI technology in some way, with 25% using AI to complete assignments for them. The most common reason cited for using AI to cheat was pressure to get high grades. 97% of students wanted some form of action from schools on the threat to academic integrity caused by AI, with the most popular options being clearer policies and more education about ethical uses of AI. In September 2025, The Atlantic published an op-ed from a high school senior arguing that the normalization of AI cheating was eroding critical thinking, academic integrity, creativity, and the shared student experience.

  Read more →
• 

  Data thinking

  Data Thinking is a framework that integrates data science with the design process. It combines computational thinking, statistical thinking, and domain-specific knowledge to guide the development of data-driven solutions in product development. The framework is used to explore, design, develop, and validate solutions, with a focus on user experience and data analytics, including data collection and interpretation The framework aims to apply data literacy and inform decision-making through data-driven insights. == Major components == According to "Computational thinking in the era of data science": Data thinking involves understanding that solutions require both data-driven and domain-knowledge-driven rules. Data thinking evaluates whether data accurately represents real-life scenarios and improves data collection where necessary. The framework highlights the importance of preserving domain-specific meaning during data analysis. Data thinking incorporates statistical and logical analysis to identify patterns and irregularities. Data thinking involves testing solutions in real-life contexts and iteratively improving models based on new data. The process requires evaluating problems from multiple abstraction levels and understanding the potential for biases in generalizations. == Major phases == === Strategic context and risk analysis === Analyzing the broader digital strategy and assessing risks and opportunities is a common step before beginning a project. Techniques like coolhunting, trend analysis, and scenario planning can be used to assist with this. === Ideation and exploration === In this phase, focus areas are identified, and use cases are developed by integrating organizational goals, user needs, and data requirements. Design thinking methods, such as personas and customer journey mapping, are applied. === Prototyping === A proof of concept is created to test feasibility and refine solutions through iterative evaluation to optimize for effective performance. === Implementation and monitoring === Solutions are tested and monitored for performance and continual improvement. == Implementing Data Thinking == The following resources explain more about data thinking and its applications: "Data Thinking: Framework for data-based solutions" by StackFuel "What is Data Thinking? A modern approach to designing a data strategy" by Mantel Group "Data Science Thinking" by SpringerLink These sources provide detailed insights into the methodology, phases, and benefits of adopting Data Thinking in organizational processes.

  Read more →
• 

  Kruskal count

  The Kruskal count (also known as Kruskal's principle, Dynkin–Kruskal count, Dynkin's counting trick, Dynkin's card trick, coupling card trick or shift coupling) is a probabilistic concept originally demonstrated by the Russian mathematician Evgenii Borisovich Dynkin in the 1950s or 1960s discussing coupling effects and rediscovered as a card trick by the American mathematician Martin David Kruskal in the early 1970s as a side-product while working on another problem. It was published by Kruskal's friend Martin Gardner and magician Karl Fulves in 1975. This is related to a similar trick published by magician Alexander F. Kraus in 1957 as Sum total and later called Kraus principle. Besides uses as a card trick, the underlying phenomenon has applications in cryptography, code breaking, software tamper protection, code self-synchronization, control-flow resynchronization, design of variable-length codes and variable-length instruction sets, web navigation, object alignment, and others. == Card trick == The trick is performed with cards, but is more a magical-looking effect than a conventional magic trick. The magician has no access to the cards, which are manipulated by members of the audience. Thus sleight of hand is not possible. Rather the effect is based on the mathematical fact that the output of a Markov chain, under certain conditions, is typically independent of the input. A simplified version using the hands of a clock performed by David Copperfield is as follows. A volunteer picks a number from one to twelve and does not reveal it to the magician. The volunteer is instructed to start from 12 on the clock and move clockwise by a number of spaces equal to the number of letters that the chosen number has when spelled out. This is then repeated, moving by the number of letters in the new number. The output after three or more moves does not depend on the initially chosen number and therefore the magician can predict it.

  Read more →
• 

  SIPRNet

  The Secret Internet Protocol Router Network (SIPRNet) is "a system of interconnected computer networks used by the U.S. Department of Defense and the U.S. Department of State to transmit classified information (up to and including information classified SECRET) by packet switching over the 'completely secure' environment". It also provides services such as hypertext document access and electronic mail. SIPRNet is a component of the Defense Information Systems Network. Other components handle communications with other security needs, such as the NIPRNet, which is used for nonsecure communications, and the Joint Worldwide Intelligence Communications System (JWICS), which is used for Top Secret communications. == Access == According to the U.S. Department of State Web Development Handbook, domain structure and naming conventions are the same as for the open internet, except for the addition of a second-level domain, like, e.g., "sgov" between state and gov: openforum.state.sgov.gov. Files originating from SIPRNet are marked by a header tag "SIPDIS" (SIPrnet DIStribution). A corresponding second-level domain smil.mil exists for DoD users. Access is also available to a "...small pool of trusted allies, including Australia, Canada, the United Kingdom and New Zealand...". This group (including the US) is known as the Five Eyes. SIPRNet was one of the networks accessed by Chelsea Manning, convicted of leaking the video used in WikiLeaks' "Collateral Murder" release as well as the source of the US diplomatic cables published by WikiLeaks in November 2010. == Alternate names == SIPRNet and NIPRNet are referred to colloquially as SIPPERnet and NIPPERnet (or simply sipper and nipper), respectively.

  Read more →
• 

  Kernel density estimation

  In statistics, kernel density estimation (KDE) is the application of kernel smoothing for probability density estimation, i.e., a non-parametric method to estimate the probability density function of a random variable based on kernels as weights. KDE answers a fundamental data smoothing problem where inferences about the population are made based on a finite data sample. In some fields such as signal processing and econometrics it is also termed the Parzen–Rosenblatt window method, after Emanuel Parzen and Murray Rosenblatt, who are usually credited with independently creating it in its current form. One of the famous applications of kernel density estimation is in estimating the class-conditional marginal densities of data when using a naive Bayes classifier, which can improve its prediction accuracy. == Definition == Let x = ( x 1 , x 2 , x 3 , . . . ) {\displaystyle \mathbf {x} =\left(x_{1},x_{2},x_{3},...\right)} be independent and identically distributed samples drawn from some univariate distribution with an unknown density f at any given point x. We are interested in estimating the shape of this function f. Its kernel density estimator is f ^ h ( x ) = 1 n ∑ i = 1 n K h ( x − x i ) = 1 n h ∑ i = 1 n K ( x − x i h ) , {\displaystyle {\hat {f}}_{h}(x)={\frac {1}{n}}\sum _{i=1}^{n}K_{h}(x-x_{i})={\frac {1}{nh}}\sum _{i=1}^{n}K{\left({\frac {x-x_{i}}{h}}\right)},} where K is the kernel — a non-negative function — and h > 0 is a smoothing parameter called the bandwidth or simply width. A kernel with subscript h is called the scaled kernel and defined as Kh(x) = ⁠1/h⁠ K(⁠x/h⁠). Intuitively one wants to choose h as small as the data will allow; however, there is always a trade-off between the bias of the estimator and its variance. The choice of bandwidth is discussed in more detail below. A range of kernel functions are commonly used: uniform, triangular, biweight, triweight, Epanechnikov (parabolic), normal, and others. The Epanechnikov kernel is optimal in a mean square error sense, though the loss of efficiency is small for the kernels listed previously. Due to its convenient mathematical properties, the normal kernel is often used, which means K(x) = ϕ(x), where ϕ is the standard normal density function. The kernel density estimator then becomes f ^ h ( x ) = 1 n ∑ i = 1 n 1 h 2 π exp ⁡ ( − ( x − x i ) 2 2 h 2 ) , {\displaystyle {\hat {f}}_{h}(x)={\frac {1}{n}}\sum _{i=1}^{n}{\frac {1}{h{\sqrt {2\pi }}}}\exp \left({\frac {-(x-x_{i})^{2}}{2h^{2}}}\right),} where h {\displaystyle h} is the standard deviation of the sample x {\displaystyle \mathbf {x} } . The construction of a kernel density estimate finds interpretations in fields outside of density estimation. For example, in thermodynamics, this is equivalent to the amount of heat generated when heat kernels (the fundamental solution to the heat equation) are placed at each data point locations xi. Similar methods are used to construct discrete Laplace operators on point clouds for manifold learning (e.g. diffusion map). == Example == Kernel density estimates are closely related to histograms, but can be endowed with properties such as smoothness or continuity by using a suitable kernel. The diagram below based on these 6 data points illustrates this relationship: For the histogram, first, the horizontal axis is divided into sub-intervals or bins which cover the range of the data: In this case, six bins each of width 2. Whenever a data point falls inside this interval, a box of height 1/12 is placed there. If more than one data point falls inside the same bin, the boxes are stacked on top of each other. For the kernel density estimate, normal kernels with a standard deviation of 1.5 (indicated by the red dashed lines) are placed on each of the data points xi. The kernels are summed to make the kernel density estimate (solid blue curve). The smoothness of the kernel density estimate (compared to the discreteness of the histogram) illustrates how kernel density estimates converge faster to the true underlying density for continuous random variables. == Bandwidth selection == The bandwidth of the kernel is a free parameter which exhibits a strong influence on the resulting estimate. To illustrate its effect, we take a simulated random sample from the standard normal distribution (plotted at the blue spikes in the rug plot on the horizontal axis). The grey curve is the true density (a normal density with mean 0 and variance 1). In comparison, the red curve is undersmoothed since it contains too many spurious data artifacts arising from using a bandwidth h = 0.05, which is too small. The green curve is oversmoothed since using the bandwidth h = 2 obscures much of the underlying structure. The black curve with a bandwidth of h = 0.337 is considered to be optimally smoothed since its density estimate is close to the true density. An extreme situation is encountered in the limit h → 0 {\displaystyle h\to 0} (no smoothing), where the estimate is a sum of n delta functions centered at the coordinates of analyzed samples. In the other extreme limit h → ∞ {\displaystyle h\to \infty } the estimate retains the shape of the used kernel, centered on the mean of the samples (completely smooth). The most common optimality criterion used to select this parameter is the expected L2 risk function, also termed the mean integrated squared error: MISE ⁡ ( h ) = E [ ∫ ( f ^ h ( x ) − f ( x ) ) 2 d x ] {\displaystyle \operatorname {MISE} (h)=\operatorname {E} \!\left[\int \!{\left({\hat {f}}\!_{h}(x)-f(x)\right)}^{2}dx\right]} Under weak assumptions on f and K, (f is the, generally unknown, real density function), MISE ⁡ ( h ) = AMISE ⁡ ( h ) + o ( ( n h ) − 1 + h 4 ) {\displaystyle \operatorname {MISE} (h)=\operatorname {AMISE} (h)+{\mathcal {o}}{\left((nh)^{-1}+h^{4}\right)}} where o is the little o notation, and n the sample size (as above). The AMISE is the asymptotic MISE, i. e. the two leading terms, AMISE ⁡ ( h ) = R ( K ) n h + 1 4 m 2 ( K ) 2 h 4 R ( f ″ ) {\displaystyle \operatorname {AMISE} (h)={\frac {R(K)}{nh}}+{\frac {1}{4}}m_{2}(K)^{2}h^{4}R(f'')} where R ( g ) = ∫ g ( x ) 2 d x {\textstyle R(g)=\int g(x)^{2}\,dx} for a function g, m 2 ( K ) = ∫ x 2 K ( x ) d x {\textstyle m_{2}(K)=\int x^{2}K(x)\,dx} and f ″ {\displaystyle f''} is the second derivative of f {\displaystyle f} and K {\displaystyle K} is the kernel. The minimum of this AMISE is the solution to this differential equation ∂ ∂ h AMISE ⁡ ( h ) = − R ( K ) n h 2 + m 2 ( K ) 2 h 3 R ( f ″ ) = 0 {\displaystyle {\frac {\partial }{\partial h}}\operatorname {AMISE} (h)=-{\frac {R(K)}{nh^{2}}}+m_{2}(K)^{2}h^{3}R(f'')=0} or h AMISE = R ( K ) 1 / 5 m 2 ( K ) 2 / 5 R ( f ″ ) 1 / 5 n − 1 / 5 = C n − 1 / 5 {\displaystyle h_{\operatorname {AMISE} }={\frac {R(K)^{1/5}}{m_{2}(K)^{2/5}R(f'')^{1/5}}}n^{-1/5}=Cn^{-1/5}} Neither the AMISE nor the hAMISE formulas can be used directly since they involve the unknown density function f {\displaystyle f} or its second derivative f ″ {\displaystyle f''} . To overcome that difficulty, a variety of automatic, data-based methods have been developed to select the bandwidth. Several review studies have been undertaken to compare their efficacies, with the general consensus that the plug-in selectors and cross validation selectors are the most useful over a wide range of data sets. Substituting any bandwidth h which has the same asymptotic order n−1/5 as hAMISE into the AMISE gives that AMISE(h) = O(n−4/5), where O is the big O notation. It can be shown that, under weak assumptions, there cannot exist a non-parametric estimator that converges at a faster rate than the kernel estimator. Note that the n−4/5 rate is slower than the typical n−1 convergence rate of parametric methods. If the bandwidth is not held fixed, but is varied depending upon the location of either the estimate (balloon estimator) or the samples (pointwise estimator), this produces a particularly powerful method termed adaptive or variable bandwidth kernel density estimation. Bandwidth selection for kernel density estimation of heavy-tailed distributions is relatively difficult. === A rule-of-thumb bandwidth estimator === If Gaussian basis functions are used to approximate univariate data, and the underlying density being estimated is Gaussian, the optimal choice for h (that is, the bandwidth that minimises the mean integrated squared error) is: h = ( 4 σ ^ 5 3 n ) 1 / 5 ≈ 1.06 σ ^ n − 1 / 5 , {\displaystyle h={\left({\frac {4{\hat {\sigma }}^{5}}{3n}}\right)}^{1/5}\approx 1.06\,{\hat {\sigma }}\,n^{-1/5},} An h {\displaystyle h} value is considered more robust when it improves the fit for long-tailed and skewed distributions or for bimodal mixture distributions. This is often done empirically by replacing the standard deviation σ ^ {\displaystyle {\hat {\sigma }}} by the parameter A {\displaystyle A} below: A = min ( σ ^ , I Q R 1.34 ) {\displaystyle A=\min \left({\hat {\sigma }},{\frac {\mathrm {IQR} }{1.34}}\right)} where IQR is the

  Read more →
• 

  Blacker (security)

  Blacker (styled BLACKER) is a U.S. Department of Defense computer network security project designed to achieve A1 class ratings (very high assurance) of the Trusted Computer System Evaluation Criteria (TCSEC). The first Blacker program began in the late 1970s, with a follow-on eventually producing fielded devices in the late 1980s. It was the first secure system with trusted end-to-end encryption on the United States' Defense Data Network. The project was implemented by SDC (software), and Burroughs (hardware), and after their merger, by the resultant company Unisys.

  Read more →
• 

  Dynamic knowledge repository

  The dynamic knowledge repository (DKR) is a concept developed by Douglas C. Engelbart as a primary strategic focus for allowing humans to address complex problems. He has proposed that a DKR will enable us to develop a collective IQ greater than any individual's IQ. References and discussion of Engelbart's DKR concept are available at the Doug Engelbart Institute. == Definition == A knowledge repository is a computerized system that systematically captures, organizes and categorizes an organization's knowledge. The repository can be searched and data can be quickly retrieved. The effective knowledge repositories include factual, conceptual, procedural and meta-cognitive techniques. The key features of knowledge repositories include communication forums. A knowledge repository can take many forms to "contain" the knowledge it holds. A customer database is a knowledge repository of customer information and insights – or electronic explicit knowledge. A Library is a knowledge repository of books – physical explicit knowledge. A community of experts is a knowledge repository of tacit knowledge or experience. The nature of the repository only changes to contain/manage the type of knowledge it holds. A repository (as opposed to an archive) is designed to get knowledge out. It should therefore have some rules of structure, classification, taxonomy, record management, etc., to facilitate user engagement.

  Read more →
• 

  Social media intelligence

  Social media intelligence (SMI or SOCMINT) comprises the collective tools and solutions that allow organizations to analyze conversations, respond to synchronize social signals, and synthesize social data points into meaningful trends and analysis, based on the user's needs. Social media intelligence allows one to utilize intelligence gathering from social media sites, using both intrusive or non-intrusive means, from open and closed social networks. This type of intelligence gathering is one element of OSINT (Open- Source Intelligence). To support both the sensing and seizing of social signals at scale, organisations increasingly rely on dedicated audience intelligence platforms which combine data aggregation, NLP-driven analysis, and cross-platform monitoring. The term 'Social Media Intelligence' was coined in a 2012 paper written by Sir David Omand, Jamie Bartlett and Carl Miller for the Centre for the Analysis of Social Media, at the London-based think tank, Demos. The authors argued that social media is now an important part of intelligence and security work, but that technological, analytical, and regulatory changes are needed before it can be considered a powerful new form of intelligence, including amendments to the United Kingdom Regulation of Investigatory Powers Act 2000. Given the dynamic evolution of social media and social media monitoring, our current understanding of how social media monitoring can help organizations create business value is inadequate. As a result, there is a need to study how organizations can (a) extract and analyze social media data related to their business (Sensing), and (b) utilize external intelligence gained from social media monitoring for specific business initiatives (Seizing). == Governmental use == In Thailand, the Technology Crime Suppression Division not only employs a 30-person team to scrutinize social media for content deemed disrespectful to the monarchy, known as lèse-majesté but also encourages citizens to report such content. Particularly targeting the youth, they run a "Cyber Scout" program where participants are rewarded for reporting individuals posting material perceived as detrimental to the monarchy. Instances in Israel involve the arrest of Palestinians by the police for their social media posts. An example includes a 15-year-old girl who posted a Facebook status with the words "forgive me," raising suspicions among Israeli authorities that she might be planning an attack. In Egypt, a leaked 2014 call for tender from the Ministry of Interior reveals efforts to procure a social media monitoring system to identify leading figures and prevent protests before they occur. In the United States, ZeroFOX faced criticism for sharing a report with Baltimore officials showcasing how their social media monitoring tool could track riots following Freddie Gray's funeral. The report labeled 19 individuals, including two prominent figures from the #BlackLivesMatter movement, as "threat actors." In the UK, the Association of Chief Police Officers of England, Wales, and Northern Ireland emphasized the significance of social media in intelligence gathering during anti-fracking protests in 2011. Social media analysis closely monitored protests against the badger cull in 2013, with a 2013 report revealing a team of 17 officers in the National Domestic Extremism Unit scanning public tweets, YouTube videos, Facebook profiles, and other online content from UK citizens. == Effects on political opinion == During the 2016 United States presidential election, the Senate Intelligence Committee released reports containing information about Russia’s use of troll farms to mislead black voters about voting. Also, German researchers in 2010 analyzed Twitter messages regarding the German federal election concluding that Twitter played a role in leading users to a specific political opinion. In a broad sense, social media refers to a conversational, distributed mode of content generation, dissemination, and communication among communities. Different from broadcast-based traditional and industrial media, social media has torn down the boundaries between authorship and readership, while the information consumption and dissemination process is becoming intrinsically intertwined with the process of generating and sharing information. An example of how SOCMINT is used to affect political opinions is the Cambridge Analytica Scandal. Cambridge Analytica was a company that purchased data from Facebook about its users without the consent or knowledge of Americans. They used this data to build a "psychological warfare tool" to persuade US voters to elect Donald Trump as president in the 2016 election. Christopher Wylie, the whistleblower, reported that personal information was taken in early 2014, and used to build a system that could target US voters with personalized pollical advertisements. More than 50 million individuals' data was exploited and manipulated. == Law enforcement == In September of 2023, the Philadelphia Police Department began using social media to track and stay one step ahead of criminal activity to stop meetups and potential robberies. This new approach has made officers utilize another tool in their field by being able to find new information as quickly as possible. Law enforcement agencies worldwide are increasingly employing social media intelligence to enhance their capabilities in both crime prevention and investigation. By analyzing publicly available data from social platforms such as Facebook, Twitter, and Instagram, police can track criminal activities, identify suspects, and even prevent potential crimes before they occur. For instance, the FBI utilizes SOCMINT to monitor threats and investigate criminal activities, including analyzing posts, images, and videos that might signal illegal activities or security concerns. == Marketing == SOCMINT collects data from both organizations and people on an individual level. It has a variety of different purposes, and though its main goal is to improve national security advancements, there are several other benefits as well. This intelligence can identify patterns, predict trends, gather information in current time, etc. In addition, these aspects have allowed for both improvement within businesses and help for law enforcement. Artificial Social Networking Intelligence (ASNI) refers to the application of artificial intelligence within social networking services and social media platforms. It encompasses various technologies and techniques used to automate, personalize, enhance, improve, and synchronize user's interactions and experiences within social networks. ASNI is expected to evolve rapidly, influencing how we interact online and shaping their digital experiences. Transparency, ethical considerations, media influence bias, and user control over data will be crucial to ensure responsible development and positive impact. Google provides many free services and has built an entire media brand with its vast variety of products. Along with data collection, Google also owns two advertising services, Google Ads, and Google AdSense. Surprisingly, most of its revenue comes from advertising, not direct sales of its services or products. Google makes money by selling advertising services to advertisers. They provide ad space to websites on Google, and target ads to consumers of Google services and products. Google can market ads using SOCMINT to collect data from its users and generate revenue. Research shows that various social media platforms on the Internet such as Twitter, Tumblr (micro-blogging websites), Facebook (a popular social networking website), YouTube (largest video sharing and hosting website), Blogs and discussion forums are being misused by extremist groups for spreading their beliefs and ideologies, promoting radicalization, recruiting members and creating online virtual communities sharing a common agenda. Popular microblogging websites such as Twitter are being used as a real-time platform for information sharing and communication during the planning and mobilization of civil unrest-related events.

  Read more →
• 

  Adversarial machine learning

  Adversarial machine learning is the study of the attacks on machine learning algorithms, and of the defenses against such attacks. Machine learning techniques are mostly designed to work on specific problem sets, under the assumption that the training and test data are generated from the same statistical distribution (IID). However, this assumption is often violated in practical high-stake applications, where users may intentionally supply fabricated data that violates the statistical assumption. Most common attacks in adversarial machine learning include evasion attacks, data poisoning attacks, Byzantine attacks and model extraction. == History == At the MIT Spam Conference in January 2004, John Graham-Cumming showed that a machine-learning spam filter could be used to defeat another machine-learning spam filter by automatically learning which words to add to a spam email to get the email classified as not spam. In 2004, Nilesh Dalvi and others noted that linear classifiers used in spam filters could be defeated by simple "evasion attacks" as spammers inserted "good words" into their spam emails. (Around 2007, some spammers added random noise to fuzz words within "image spam" in order to defeat OCR-based filters.) In 2006, Marco Barreno and others published "Can Machine Learning Be Secure?", outlining a broad taxonomy of attacks. As late as 2013 many researchers continued to hope that non-linear classifiers (such as support vector machines and neural networks) might be robust to adversaries, until Battista Biggio and others demonstrated the first gradient-based attacks on such machine-learning models (2012–2013). In 2012, deep neural networks began to dominate computer vision problems; starting in 2014, Christian Szegedy and others demonstrated that deep neural networks could be fooled by adversaries, again using a gradient-based attack to craft adversarial perturbations. Further work would show that adversarial attacks are harder to produce in uncontrolled environments, due to the different environmental constraints that cancel out the effect of noise. For example, any small rotation or slight illumination on an adversarial image can destroy the adversariality. In addition, researchers such as Google Brain's Nick Frosst point out that it is much easier to make self-driving cars miss stop signs by physically removing the sign itself, rather than creating adversarial examples. Frosst also believes that the adversarial machine learning community incorrectly assumes models trained on a certain data distribution will also perform well on a completely different data distribution. He suggests that a new approach to machine learning should be explored, and is currently working on a unique neural network that has characteristics more similar to human perception than state-of-the-art approaches. While adversarial machine learning continues to be heavily rooted in academia, large tech companies such as Google, Microsoft, and IBM have begun curating documentation and open source code bases to allow others to concretely assess the robustness of machine learning models and minimize the risk of adversarial attacks. === Examples === Examples include attacks in spam filtering, where spam messages are obfuscated through the misspelling of "bad" words or the insertion of "good" words; attacks in computer security, such as obfuscating malware code within network packets or modifying the characteristics of a network flow to mislead intrusion detection; attacks in biometric recognition where fake biometric traits may be exploited to impersonate a legitimate user; or to compromise users' template galleries that adapt to updated traits over time. Researchers showed that by changing only one-pixel it was possible to fool deep learning algorithms. Others 3-D printed a toy turtle with a texture engineered to make Google's object detection AI classify it as a rifle regardless of the angle from which the turtle was viewed. Creating the turtle required only low-cost commercially available 3-D printing technology. A machine-tweaked image of a dog was shown to look like a cat to both computers and humans. A 2019 study reported that humans can guess how machines will classify adversarial images. Researchers discovered methods for perturbing the appearance of a stop sign such that an autonomous vehicle classified it as a merge or speed limit sign. A data poisoning filter called Nightshade was released in 2023 by researchers at the University of Chicago. It was created for use by visual artists to put on their artwork to corrupt the data set of text-to-image models, which usually scrape their data from the internet without the consent of the image creator. McAfee attacked Tesla's former Mobileye system, fooling it into driving 50 mph over the speed limit, simply by adding a two-inch strip of black tape to a speed limit sign. Adversarial patterns on glasses or clothing designed to deceive facial-recognition systems or license-plate readers, have led to a niche industry of "stealth streetwear". An adversarial attack on a neural network can allow an attacker to inject algorithms into the target system. Researchers can also create adversarial audio inputs to disguise commands to intelligent assistants in benign-seeming audio; a parallel literature explores human perception of such stimuli. Clustering algorithms are used in security applications. Malware and computer virus analysis aims to identify malware families, and to generate specific detection signatures. In the context of malware detection, researchers have proposed methods for adversarial malware generation that automatically craft binaries to evade learning-based detectors while preserving malicious functionality. Optimization-based attacks such as GAMMA use genetic algorithms to inject benign content (for example, padding or new PE sections) into Windows executables, framing evasion as a constrained optimization problem that balances misclassification success with the size of the injected payload and showing transferability to commercial antivirus products. Complementary work uses generative adversarial networks (GANs) to learn feature-space perturbations that cause malware to be classified as benign; Mal-LSGAN, for instance, replaces the standard GAN loss with a least-squares objective and modified activation functions to improve training stability and produce adversarial malware examples that substantially reduce true positive rates across multiple detectors. == Challenges in applying machine learning to security == Researchers have observed that the constraints under which machine-learning techniques function in the security domain are different from those of common benchmark domains. Security data may change over time, include mislabeled samples, or reflect adversarial behavior, which complicates evaluation and reproducibility. === Data collection issues === Security datasets vary across formats, including binaries, network traces, and log files. Studies have reported that the process of converting these sources into features can introduce bias or inconsistencies. In addition, time-based leakage can occur when related malware samples are not properly separated across training and testing splits, which may lead to overly optimistic results. === Labeling and ground truth challenges === Malware labels are often unstable because different antivirus engines may classify the same sample in conflicting ways. Ceschin et al. note that families may be renamed or reorganized over time, causing further discrepancies in ground truth and reducing the reliability of benchmarks. === Concept drift === Because malware creators continuously adapt their techniques, the statistical properties of malicious samples also change. This form of concept drift has been widely documented and may reduce model performance unless systems are updated regularly or incorporate mechanisms for incremental learning. === Feature robustness === Researchers differentiate between features that can be easily manipulated and those that are more resistant to modification. For example, simple static attributes, such as header fields, may be altered by attackers, while structural features, such as control-flow graphs, are generally more stable but computationally expensive to extract. === Class imbalance === In realistic deployment environments, the proportion of malicious samples can be extremely low, ranging from 0.01% to 2% of total data. This unbalanced distribution causes models to develop a bias towards the majority class, achieving high accuracy but failing to identify malicious samples. Prior approaches to this problem have included both data-level solutions and sequence-specific models. Methods like n-gram and Long Short-Term Memory (LSTM) networks can model sequential data, but their performance has been shown to decline significantly when malware samples are realistically proportioned in the training set, demonstrating the limitations in

  Read more →
• 

  Hardware random number generator

  In computing, a hardware random number generator (HRNG), true random number generator (TRNG), non-deterministic random bit generator (NRBG), or physical random number generator is a device that generates random numbers from a physical process capable of producing entropy, unlike a pseudorandom number generator (PRNG) that utilizes a deterministic algorithm and non-physical nondeterministic random bit generators that do not include hardware dedicated to generation of entropy. Many natural phenomena generate low-level, statistically random "noise" signals, including thermal and shot noise, jitter and metastability of electronic circuits, Brownian motion, and atmospheric noise. Researchers also used the photoelectric effect, involving a beam splitter, other quantum phenomena, and even nuclear decay (due to practical considerations the latter, as well as the atmospheric noise, is not viable except for fairly restricted applications or online distribution services). While "classical" (non-quantum) phenomena are not truly random, an unpredictable physical system is usually acceptable as a source of randomness, so the qualifiers "true" and "physical" are used interchangeably. A hardware random number generator is expected to output near-perfect random numbers ("full entropy"). A physical process usually does not have this property, and a practical TRNG typically includes a few blocks: a noise source that implements the physical process producing the entropy. Usually this process is analog, so a digitizer is used to convert the output of the analog source into a binary representation; a conditioner (randomness extractor) that improves the quality of the random bits; health tests. TRNGs are mostly used in cryptographical algorithms that get completely broken if the random numbers have low entropy, so the testing functionality is usually included. Hardware random number generators generally produce only a limited number of random bits per second. In order to increase the available output data rate, they are often used to generate the "seed" for a faster PRNG. PRNG also helps with the noise source "anonymization" (whitening out the noise source identifying characteristics) and entropy extraction. With a proper PRNG algorithm selected (cryptographically secure pseudorandom number generator, CSPRNG), the combination can satisfy the requirements of Federal Information Processing Standards and Common Criteria standards. == Uses == Hardware random number generators can be used in any application that needs randomness. However, in many scientific applications additional cost and complexity of a TRNG (when compared with pseudo random number generators) provide no meaningful benefits. TRNGs have additional drawbacks for data science and statistical applications: impossibility to re-run a series of numbers unless they are stored, reliance on an analog physical entity can obscure the failure of the source. The TRNGs therefore are primarily used in the applications where their unpredictability and the impossibility to re-run the sequence of numbers are crucial to the success of the implementation: in cryptography and gambling machines. === Cryptography === The major use for hardware random number generators is in the field of data encryption, for example to create random cryptographic keys and nonces needed to encrypt and sign data. In addition to randomness, there are at least two additional requirements imposed by the cryptographic applications: forward secrecy guarantees that the knowledge of the past output and internal state of the device should not enable the attacker to predict future data; backward secrecy protects the "opposite direction": knowledge of the output and internal state in the future should not divulge the preceding data. A typical way to fulfill these requirements is to use a TRNG to seed a cryptographically secure pseudorandom number generator. == History == Physical devices were used to generate random numbers for thousands of years, primarily for gambling. Dice in particular have been known for more than 5000 years (found on locations in modern Iraq and Iran), and flipping a coin (thus producing a random bit) dates at least to the times of ancient Rome. The first documented use of a physical random number generator for scientific purposes was by Francis Galton (1890). He devised a way to sample a probability distribution using a common gambling die. In addition to the top digit, Galton also looked at the face of a die closest to him, thus creating 64 = 24 outcomes (about 4.6 bits of randomness). Kendall and Babington-Smith (1938) used a fast-rotating 10-sector disk that was illuminated by periodic bursts of light. The sampling was done by a human who wrote the number under the light beam onto a pad. The device was utilized to produce a 100,000-digit random number table (at the time such tables were used for statistical experiments, like PRNG nowadays). On 29 April 1947, the RAND Corporation began generating random digits with an "electronic roulette wheel", consisting of a random frequency pulse source of about 100,000 pulses per second gated once per second with a constant frequency pulse and fed into a five-bit binary counter. Douglas Aircraft built the equipment, implementing Cecil Hasting's suggestion (RAND P-113) for a noise source (most likely the well known behavior of the 6D4 miniature gas thyratron tube, when placed in a magnetic field). Twenty of the 32 possible counter values were mapped onto the 10 decimal digits and the other 12 counter values were discarded. The results of a long run from the RAND machine, filtered and tested, were converted into a table, which originally existed only as a deck of punched cards, but was later published in 1955 as a book, 50 rows of 50 digits on each page (A Million Random Digits with 100,000 Normal Deviates). The RAND table was a significant breakthrough in delivering random numbers because such a large and carefully prepared table had never before been available. It has been a useful source for simulations, modeling, and for deriving the arbitrary constants in cryptographic algorithms to demonstrate that the constants had not been selected maliciously ("nothing up my sleeve numbers"). Since the early 1950s, research into TRNGs has been highly active, with thousands of research works published and about 2000 patents granted by 2017. == Physical phenomena with random properties == Multiple different TRNG designs were proposed over time with a large variety of noise sources and digitization techniques ("harvesting"). However, practical considerations (size, power, cost, performance, robustness) dictate the following desirable traits: use of a commonly available inexpensive silicon process; exclusive use of digital design techniques. This allows an easier system-on-chip integration and enables the use of FPGAs; compact and low-power design. This discourages use of analog components (e.g., amplifiers); mathematical justification of the entropy collection mechanisms. Stipčević & Koç in 2014 classified the physical phenomena used to implement TRNG into four groups: electrical noise; free-running oscillators; chaos; quantum effects. === Electrical noise-based RNG === Noise-based RNGs generally follow the same outline: the source of a noise generator is fed into a comparator. If the voltage is above threshold, the comparator output is 1, otherwise 0. The random bit value is latched using a flip-flop. Sources of noise vary and include: Johnson–Nyquist noise ("thermal noise"); Zener noise; avalanche breakdown. The drawbacks of using noise sources for an RNG design are: noise levels are hard to control, they vary with environmental changes and device-to-device; calibration processes needed to ensure a guaranteed amount of entropy are time-consuming; noise levels are typically low, thus the design requires power-hungry amplifiers. The sensitivity of amplifier inputs enables manipulation by an attacker; circuitry located nearby generates a lot of non-random noise thus lowering the entropy; a proof of randomness is near-impossible as multiple interacting physical processes are involved. === Chaos-based RNG === The idea of chaos-based noise stems from the use of a complex system that is hard to characterize by observing its behavior over time. For example, lasers can be put into (undesirable in other applications) chaos mode with chaotically fluctuating power, with power detected using a photodiode and sampled by a comparator. The design can be quite small, as all photonics elements can be integrated on-chip. Stipčević & Koç characterize this technique as "most objectionable", mostly due to the fact that chaotic behavior is usually controlled by a differential equation and no new randomness is introduced, thus there is a possibility of the chaos-based TRNG producing a limited subset of possible output strings. === Free-running oscillators-based RNG === The TRNGs based on a free-running oscilla

  Read more →
• 

  Merit Network

  Merit Network, Inc., is a nonprofit member-governed organization providing high-performance computer networking and related services to educational, government, health care, and nonprofit organizations, primarily in Michigan. Created in 1966, Merit operates the longest running regional computer network in the United States. == Organization == Created in 1966 as the Michigan Educational Research Information Triad by Michigan State University (MSU), the University of Michigan (U-M), and Wayne State University (WSU), Merit was created to investigate resource sharing by connecting the mainframe computers at these three Michigan public research universities. Merit's initial three node packet-switched computer network was operational in October 1972 using custom hardware based on DEC PDP-11 minicomputers and software developed by the Merit staff and the staffs at the three universities. Over the next dozen years the initial network grew as new services such as dial-in terminal support, remote job submission, remote printing, and file transfer were added; as gateways to the national and international Tymnet, Telenet, and Datapac networks were established, as support for the X.25 and TCP/IP protocols was added; as additional computers such as WSU's MVS system and the UM's electrical engineering's VAX running UNIX were attached; and as new universities became Merit members. Merit's involvement in national networking activities started in the mid-1980s with connections to the national supercomputing centers and work on the 56 kbit/s National Science Foundation Network (NSFNET), the forerunner of today's Internet. From 1987 until April 1995, Merit re-engineered and managed the NSFNET backbone service. MichNet, Merit's regional network in Michigan was attached to NSFNET and in the early 1990s Merit began extending "the Internet" throughout Michigan, offering both direct connect and dial-in services, and upgrading the statewide network from 56 kbit/s to 1.5 Mbit/s, and on to 45, 155, 622 Mbit/s, and eventually 1 and 10 Gbit/s. In 2003 Merit began its transition to a facilities based network, using fiber optic facilities that it shares with its members, that it purchases or leases under long-term agreements, or that it builds. In addition to network connectivity services, Merit offers a number of related services within Michigan and beyond, including: Internet2 connectivity, VPN, Network monitoring, Voice over IP (VOIP), Cloud storage, E-mail, Domain Name, Network Time, VMware and Zimbra software licensing, Colocation, and professional development seminars, workshops, classes, conferences, and meetings. == History == === Creating the network: 1966 to 1973 === The Michigan Educational Research Information Triad (MERIT) was formed in the fall of 1966 by Michigan State University (MSU), University of Michigan (U-M), and Wayne State University (WSU). More often known as the Merit Computer Network or simply Merit, it was created to design and implement a computer network connecting the mainframe computers at the universities. In the fall of 1969, after funding for the initial development of the network had been secured, Bertram Herzog was named director for MERIT. Eric Aupperle was hired as senior engineer, and was charged with finding hardware to make the network operational. The National Science Foundation (NSF) and the State of Michigan provided the initial funding for the network. In June 1970, the Applied Dynamics Division of Reliance Electric in Saline, Michigan was contracted to build three Communication Computers or CCs. Each would consist of a Digital Equipment Corporation (DEC) PDP-11 computer, dataphone interfaces, and interfaces that would attach them directly to the mainframe computers. The cost was to be slightly less than the $300,000 ($2,487,100, adjusted for inflation) originally budgeted. Merit staff wrote the software that ran on the CCs, while staff at each of the universities wrote the mainframe software to interface to the CCs. The first completed connection linked the IBM S/360-67 mainframe computers running the Michigan Terminal System at WSU and U-M, and was publicly demonstrated on December 14, 1971. The MSU node was completed in October 1972, adding a CDC 6500 mainframe running Scope/Hustler. The network was officially dedicated on May 15, 1973. === Expanding the network: 1974 to 1985 === In 1974, Herzog returned to teaching in the University of Michigan's Industrial Engineering Department, and Aupperle was appointed as director. Use of the all uppercase name "MERIT" was abandoned in favor of the mixed case "Merit". The first network connections were host to host interactive connections which allowed person to remote computer or local computer to remote computer interactions. To this, terminal to host connections, batch connections (remote job submission, remote printing, batch file transfer), and interactive file copy were added. And, in addition to connecting to host computers over custom hardware interfaces, the ability to connect to hosts or other networks over groups of asynchronous ports and via X.25 were added. Merit interconnected with Telenet (later SprintNet) in 1976 to give Merit users dial-in access from locations around the United States. Dial-in access within the U.S. and internationally was further expanded via Merit's interconnections to Tymnet, ADP's Autonet, and later still the IBM Global Network as well as Merit's own expanding network of dial-in sites in Michigan, New York City, and Washington, D.C. In 1978, Western Michigan University (WMU) became the fourth member of Merit (prompting a name change, as the acronym Merit no longer made sense as the group was no longer a triad). To expand the network, the Merit staff developed new hardware interfaces for the Digital PDP-11 based on printed circuit technology. The new system became known as the Primary Communications Processor (PCP), with the earliest PCPs connecting a PDP-10 located at WMU and a DEC VAX running UNIX at U-M's Electrical Engineering department. A second hardware technology initiative in 1983 produced the smaller Secondary Communication Processors (SCP) based on DEC LSI-11 processors. The first SCP was installed at the Michigan Union in Ann Arbor, creating UMnet, which extended Merit's network connectivity deeply into the U-M campus. In 1983 Merit's PCP and SCP software was enhanced to support TCP/IP and Merit interconnected with the ARPANET. === National networking, NSFNET, and the Internet: 1986 to 1995 === In 1986 Merit engineered and operated leased lines and satellite links that allowed the University of Michigan to access the supercomputing facilities at Pittsburgh, San Diego, and NCAR. In 1987, Merit, IBM and MCI submitted a winning proposal to NSF to implement a new NSFNET backbone network. The new NSFNET backbone network service began July 1, 1988. It interconnected supercomputing centers around the country at 1.5 megabits per second (T1), 24 times faster than the 56 kilobits-per-second speed of the previous network. The NSFNET backbone grew to link scientists and educators on university campuses nationwide and connect them to their counterparts around the world. The NSFNET project caused substantial growth at Merit, nearly tripling the staff and leading to the establishment of a new 24-hour Network Operations Center at the U-M Computer Center. In September 1990 in anticipation of the NSFNET T3 upgrade and the approaching end of the 5-year NSFNET cooperative agreement, Merit, IBM, and MCI formed Advanced Network and Services (ANS), a new non-profit corporation with a more broadly based Board of Directors than the Michigan-based Merit Network. Under its cooperative agreement with NSF, Merit remained ultimately responsible for the operation of NSFNET, but subcontracted much of the engineering and operations work to ANS. In 1991 the NSFNET backbone service was expanded to additional sites and upgraded to a more robust 45 Mbit/s (T3) based network. The new T3 backbone was named ANSNet and provided the physical infrastructure used by Merit to deliver the NSFNET Backbone Service. On April 30, 1995, the NSFNET project came to an end, when the NSFNET backbone service was decommissioned and replaced by a new Internet architecture with commercial Internet service providers (ISPs) interconnected at Network Access Points provided by multiple providers across the country. === Bringing the Internet to Michigan: 1985 to 2001 === During the 1980s, Merit Network grew to serve eight member universities, with Oakland University joining in 1985 and Central Michigan University, Eastern Michigan University, and Michigan Technological University joining in 1987. In 1990, Merit's board of directors formally changed the organization's name to Merit Network, Inc., and created the name MichNet to refer to Merit's statewide network. The board also approved a staff proposal to allow organizations other than publicly supported universities, referred to as aff

  Read more →