Seção Tecnologia - Testei

Snakey Webcam Flash Drive

Testei em Windows XP este produto

A qualidade de imagem é relativamente boa para transmissão via WEB, via messenger, Yahoo messenger.
A dificuldade é encontrar angulo que de visão ambiental. A cobra metálica permite múltiplos ajustes de posição mas mesmo assim é difícil encontrar um boa posição. O detalhe é possuir dois modos de operação, Quando camera não permite acesso de storage, e quando storage não permite uso de camera. Ma so produto é interessante por ser um dois em um, muito vantajoso.




Yet Another Flash Drive, this one with a built-in sensor designed specifically for use as a webcam, plus all the usual USB 1.1 (crappy) data rates. Plus, it has a snapshot button for when you're carrying it around in your pocket and need to take a cameraphone-quality VGA photograph. No idea where you could buy it retail, though; it was sent to me in one of those 'would Gizmodo like to buy 2000 units of our golden product sample?' spams I get every day, this one from Wide Tech Ltd. of Shin-lung Rd. Sec.3, Taipei, Taiwan.

Posted by Felipe Machado em Sábado, Maio 22, 2004

Comentários:

Show de notebook



Posted by Felipe Machado em Sábado, Maio 22, 2004

Comentários:

A paranóia americana continua

'Database firm listed 120,000 'likely terrorists''
The ACLU called the data mining effort an invasion of privacy

MAY 21, 2004 (REUTERS) - The company that runs the multistate Matrix law enforcement database gave the U.S. government a list of 120,000 people who scored high on a computer profile it said was designed to identify likely terrorists, a civil liberties group said yesterday.
The Florida company that created the list, Seisint Inc., said in government documents obtained by the American Civil Liberties Union that the "High Terrorism Factor" list had led to scores of arrests. Seisint created the list shortly after the Sept. 11, 2001, attacks, apparently on its own, the ACLU said.

The ACLU has asked for a government investigation to determine who had access to the list and how the information was used. It called the data mining program a chilling invasion of privacy that allows police to investigate millions of law-abiding citizens without their knowledge.

"People on that list ought to be concerned," said Barry Steinhardt, director of the ACLU's privacy and technology program. "Just being associated with a terrorist list is going to make people's lives miserable."

Law enforcement officials involved with the program have said the terrorist quotient was abandoned when Matrix was developed. But the ACLU said the documents it received through freedom-of-information requests contained nothing to indicate that, and it added that the terrorist profiling feature was "a sort of central selling point of the program."

Seisint didn¿t return repeated calls for comment. The privately held company in Boca Raton, Fla., creates data mining products that rapidly analyze billions of computerized records to extract useful information.

Based on profiles of the actual Sept. 11 hijackers, the terrorism quotient program examined criteria such as age, gender, ethnicity, driving records, pilot licensing records, credit history and proximity to addresses and phone numbers previously linked to criminal activity, the ACLU said.
Seisint gave the list of the highest scorers to federal authorities and claimed that the technology led to scores of arrests, according to documents obtained by the ACLU.
Last year, Seisint won a contract to run the Multistate Anti-Terrorist Information Exchange, a pilot program known as Matrix that lets law enforcement agencies share information and rapidly sort through billions of computerized records to track criminals.

In addition to using public documents such as criminal records and motor vehicle records, Matrix includes records such as credit histories and raw data from police investigators, which the ACLU said could contain errors that lead to false accusations. "It's a basic principle of American law, American policy, that you don't investigate people unless you have some reason to do so, some reasonable suspicion," Steinhardt said.

Matrix has already been controversial. It was launched in 2002 with 13 states that included more than half of the U.S. population, but many have dropped out because of cost and privacy concerns. Only five states -- Florida, Pennsylvania, Connecticut, Michigan and Ohio -- which contain about 19% of the U.S. population, still participate.

The Department of Homeland Security gave the program $8 million last year. One document released by the ACLU shows that the funding agreement gave DHS "managerial oversight and control" of the program.

The ACLU, which wants Congress to cut DHS funding for the Matrix program, said it was "substantially similar" to the ill-fated Total Information Awareness program created by the Pentagon to create a global data mining program to search computer records on individuals and build profiles.

News reports about the program prompted fear that America was becoming a police state, and Congress declined to fund it


Posted by Felipe Machado em Sábado, Maio 22, 2004

Comentários:

Escaping the Paper Paradigm

Excelent Article published in DM Direct Newsletter


With the Internet and the huge memory capacity of computers, there currently is more information at our fingertips than we can utilize effectively. While access to digital information already is infinite and instantaneous, more information does not equal more knowledge. To move forward in our emerging information society, it is now a matter of urgency to shed our thinking as well as our technology of the paper paradigms of the past and embrace a mind-set that fully leverages the integration potential of digital records.
Innovations in computing have revolutionized our capacity to access from boardrooms to classrooms. Those same innovations, and our ability to communicate instantly worldwide, have accelerated the growth of information beyond anything imaginable even a decade ago. At the beginning of 1992, there were less than a half-million Internet hosts. Today there are nearly 200 million. Google, for example, is facilitating more than 150 million searches per day within a collection of more than 4 billion Web sites from around the world - generating ranked lists with millions of relevant information resources for virtually any query that hide all of the relationships within and between the information resources.
Moreover, since the beginning of the current millennium, more than 20,000 petabytes (1 petabyte equals 1 billion megabytes) of digital information has been stored on magnetic, print and optical media each year (see Figure 1). This production of digital information is staggering all by itself and outright overwhelming considering that more than 80 percent of corporate data comes in the form of unstructured content that cannot be decomposed into a relational schema, as noted by Oracle and Documentum, for example.



Figure 1: Production rates of digital information from various storage media as estimated in 2000

Driven by necessity, our ability to process digital information has been expanding exponentially since the early 1970s (see Figure 2). Yet, despite the tremendous access that is afforded by the Internet and computer hardware, the vast majority of digital information is lumped into binary large objects (BLOBS). This is, at present, the most "efficient" manner in which unstructured data can be managed with existing software solutions. Even the world's largest information-technology companies have no better strategy for managing the avalanche of information than BLOBS.



From a historical perspective, the speed and impact of information's growth has been unprecedented. We are still at the very early stages of transitioning from centuries of using paper into a global society that depends on digital information. To soften this transition, we have linked the new to the suddenly outdated familiar, as has been true for other technological revolutions. After generations of horse-drawn carriages, the concept of "horsepower" is still applied to automobiles. After centuries of managing papers and other printed materials, computer "folders" are now used to organize digital records. Similarly, like the carbon copies of handwritten or typed messages, "cc" is used to "e-mail" duplicates of computer files.
Our information society should be looking toward a future horizon, not one from our past. Instead of looking for solutions to the information explosion through an archaic lens, we should focus on the possibilities inherent in digital records. The future of digital knowledge will not be realized in how we managed paper records in the past, only with more information and faster access. The future of information management is to comprehensively integrate and discover relationships within and between diverse information resources based on user-defined objectives and criteria. Yet how do we open the doors into this next stage in the evolution of information technologies?
The first step is to recognize that a principal limitation in our information society is the "paper paradigm." As much as anything else, our thinking has limited our technology knowledge and potential for knowledge discovery. We utilize folders to organize digital records. We create static tags to summarize digital records and make them searchable. We add markup language to digital records to express their hardcopy formats. The result of our adherence to the paper paradigm has been an unnecessary limitation we have placed on ourselves.
For digital records, these paper paradigm approaches are like building a library without an effective cataloguing scheme. Consequently, to access information we would need a brute-force tool - a search engine - that would stack all of the pertinent books in a large pile (just like the lists of "hits" that we generate from a digital search query). We then would need to check titles and open pages to identify information of interest. Subsequent analyses to describe relationships and trends among the books would involve manual integration - and an unbearably inefficient investment of time to turn a collection of data sources into usable insights.
Fortunately, there are widely accepted strategies (such as the Dewey Decimal System or the Library of Congress system) for tagging, relating and locating books in hierarchies of shelves on floors within libraries. However, unlike books, which are tagged once and forever based on their content, digital records have the potential to be retagged based on user-defined objectives to manage them within dynamic collections.
For these dynamic collections, various "standards" of meta data tags have emerged to describe, structure and administrate digital records. However, because of the amorphous nature of digital records, meta data tags contain numerous fields with all kinds of controlled vocabularies. In a practical sense, the effort to create these unique identifiers for every digital record makes the meta data tags just as static as tags on a book.
A further complication with meta data exists when digital records are broken into smaller pieces to increase the granularity of a collection. Under current strategies, the meta data for each record remains constant. This means that as the granularity of the data is increased, the volume of meta data necessarily grows exponentially compared to the actual content of interest as shown in Figure 3. Companion applications of markup tags, which facilitate the organization and searchability of digital records, add further to the volume of information that needs to be stored and processed. Moreover, embedding markup or format tags modifies the original authentic digital record. Despite the ongoing generation of "standards" and the industries that have emerged, current applications of meta data and markup are paper paradigm bandaids rather than long-term panaceas for managing digital records.



Figure 3: Model of the exponentially increasing volume of meta data relative to the original content caused by merely doubling the granularity, such that each information granule becomes smaller while the meta data fields are static and effectively independent of granule size.

As the growth in global information extends inevitably beyond gigabytes, terabytes and petabytes to exabytes per year, the gap between information generation and real knowledge creation is widening. The conundrum is simple enough: because we can process more information, we create more information to be processed.

The difference between combining paper documents and combining digital records is both fundamental and profound. If you take two books and put them together, you still have two books. You cannot break down the "integrity" of a paper document in order to create a new knowledge combination (and new discovery) via combinations of granularities. However, in the digital arena, it becomes possible to mix the records in potentially infinite directions and realize unexpected discoveries - to be surprised by new knowledge. The underlying advantage of digital information is the ability to automatically break information resources into pieces that can be objectively integrated into the original or new information resources.

Regarding digital records, the real opportunity is to discover knowledge by integrating information at scale based on user-defined objectives and criteria. For integration to be possible, there has to be a granularity capability that is virtually untapped today. In fact, our ability to integrate pieces of information is directly proportional to our ability to break down digital records into granules.

Granularity is the basis for integration. If you have one piece of information, there is nothing to integrate. Two pieces can be related with one intersection. Three pieces can be related with four intersections while four pieces can be related with nine intersections. In fact, the degree of integration expands geometrically with the number of granules. The more pieces you create, the greater the potential for integration and knowledge discovery.

In the paper world, when you put two books together, you still have two books. We are unable to physically break the books into chapters and sentences and words that can be dynamically integrated in a meaningful way based on user-defined objectives. The problem is that we are applying the same kind of restriction to digital records and even the Internet. Rather than integrating digital records, we aggregate them in lists and folders.

With exponentially more information to manage, such limitations have become glaring. Moreover, we continue to search for specific content within electronic documents as if they were printed pages. In doing so, we fail to realize the potential for integration - and knowledge creation - that digital records offer. The richness of the electronic world is a lot broader than anyone has yet leveraged. The overriding limitations in knowledge management are not technical, but perceptual.

The economics of more information mean that, without true integration capability, the overwhelming volume of information pieces makes it more difficult and complicated to reach decisions. The ability to integrate influences the ability to make decisions about personnel management, financial transactions, corporate relationships, marketplace strategies, national security issues, international policy, scientific research as well as average, everyday home computing.

Until we clear away the antiquated remnants of the paper paradigm and redefine our expectations of digital knowledge management strictly according to electronic parameters, we will not know what the true potential of digital records is, much less how to achieve it.

If we were not locked into the paper paradigm, how would we manage, organize, integrate and access information to achieve true insights and knowledge? I suggest that without the paper paradigm limitations, we would pursue a different approach to information processing and analysis.

Within the electronic paradigm, we would pursue an approach that would fully utilize the strengths of digital record and realize:

                - Rapid-automated ingestion and integration of digital information independent of scale, multimedia formats or sources; 

                - Modular functionality and user-defined granularity; 

                - Automated, dynamic and interoperable tagging of information granules; 

                - Comprehensive searchability and organization without markup language; and 

                - Dynamic hierarchal displays that comprehensively describe quantifiable relationships within and between information resources.

That would mean abandoning the accepted system of meta data and markup language tags in favor of an approach that facilitates unlimited granularity and automated tagging of information granules. Such an approach would allow us to organize information in such a way that we could retrieve specific information faster and within the context of its meaning as well as its location. It would eliminate the need for "folder" strategies since organization would be accomplished with individual granules as whole records that could be dynamically integrated with any or all other granules in a collection. Consequently, scale would play no role in organizing or managing the records.

Instead of relying on graphical user interfaces for locating information, we could search information granules directly and more comprehensively. In addition, we could be certain that we are working only with the authentic, untainted data granules without the possibility of contamination by meta data tags or markup language.

Clearly we are not taking full advantage of the knowledge management capabilities that today's computers and our digital records afford us. But are the current restrictions in our methods for organizing, searching and displaying digital information a function of our paradigm or the limitations of existing technology?

Paradigm precedes technological solutions. The solutions we find are the ones we look for based on our expectations of what technology can and should do. For too long, we have accepted the paper paradigm as the source of our expectations for how digital records are created, accessed and displayed. The limitations of our own expectations have allowed us to settle for only a fraction of the capability inherent in utilizing digital records. It is a paradigm we can no longer afford to rely on.

As the growth of digital information spirals beyond our ability to manage and utilize it, we will find ourselves more and more compelled to identify a new paradigm. For our future - we must have the creative capacity to dynamically integrate digital records in directions that open new dimensions of knowledge discovery for each of us individually.



Posted by Felipe Machado em Sexta-feira, Maio 21, 2004

Comentários:

Seção Gestão do Conhecimento

Qual a diferença entre conhecimento tácito, explícito e implícito?

Conhecimento explícito é o conhecimento que possuímos e de que temos consciência. É o conhecimento que somos capazes de documentar, é o conhecimento que as organizações conseguem armazenar.
O conhecimento tácito, ao contrário, é conhecimento que temos mas do qual não nos apercebemos. É conhecimento que adquirimos através da prática, da experiência, dos erros e dos sucessos. É conhecimento que não somos capazes de descrever claramente e talvez nem documentar. É conhecimento que as organizações não podem utilizar fora do horário de trabalho dos seus colaboradores.
O conhecimento implícito é um conceito mais recente e que serve para descrever conhecimento que, embora ainda não tenha sido documentado, é passível de o ser. É conhecimento que possuímos e que somos capazes de transmitir, de forma mais ou menos assistida e estruturada.

O sapo e a cobra (conhecimento implícito, tácito ou explícito ?)

Um dia, um pescador estava no seu barco no rio quando algo na água chamou a sua atenção. Ao olhar para o rio reparou numa cobra com um sapo na boca.
Debruçou-se, pegou na cobra, abriu a sua boca e atirou o sapo para o rio. "Isto é para ver se aprendes, cobra malvada", pensou o pescador para consigo.
Mas depois, pensou melhor "Na verdade o que acabei de fazer foi privar a cobra de uma refeição - talvez isso não seja muito justo". Sentindo-se culpado, olhou para o barco e reparou que nada lhe sobrara do seu almoço. "Já sei", pensou ele, "vou lhe dar cerveja."
Despejou, então, cerca de um quarto da garrafa pela garganta da cobra abaixo. Contente por ter recompensado a cobra pelo prejuízo que lhe tinha causado, o pescador colocou-a de novo na água.
Dez minutos mais tarde, foi mais uma vez acordado das suas divagações por um ligeiro bater na borda do barco. Espreitou e viu a mesma cobra, desta vez com dois sapos na boca!


Posted by Felipe Machado em Quinta-feira, Maio 20, 2004

Comentários:

Seção Besteirol da Tecnologia

Se você quer gastar muito dinheiro mas não a quantidade total para compra uma Ferrari de verdade, você agora poderá investir em uma máquina fotográfica que estará obsoleta em pouco tempo. Esta é notícia é ótima.
Olympus e a Ferrari anunciaram uma edição limitada de câmeras digitais com a marca da montadora italiana.
A máquina, que levará o nome de Ferrari Digital Model 2004, terá apenas dez mil unidades produzidas e pode ser reservada apenas pelo website da Olympus.
A câmera tem o mesmo tom de vermelho utilizado nos carros da Ferrari e será confeccionada com o material dos bólidos da Fórmula 1 - fibra de carbono. O emblema do cavallino rampante, que não poderia ficar de fora, aparece ao lado do visor em cristal líquido.



A Ferrari Digital Model 2004 apresenta também algumas peculiaridades que não deverão ser encontradas em outras máquinas comuns.
Ao ligar a câmera, a palavra "F2004" aparece no display enquanto o som de um motor de Ferrari é tocado.
E realmente fantástico, o que você acha de uma máquina fotográfica que faz ruídos de vroom de vroom?
Realmente neste mundo tem público pra tudo que se inventar de besteira associada à tecnologia. Aposto que a edição de câmeras irá se esgotar em poucos dias.


Posted by Felipe Machado em Quinta-feira, Maio 20, 2004

Comentários:

A Frase do dia

"A melhor definição de líder a que eu consegui chegar é que líder é aquele que consegue tirar das pessoas mais do que elas imaginam que são capazes de fazer"

Maria Silvia Bastos Marques.
Ex-presidente da CSN, Companhia Siderurgia Nacional

Posted by Felipe Machado em Quinta-feira, Maio 20, 2004

Comentários:

METODOLOGIAS ÁGEIS DE SOFTWARE É TEMA DE CONGRESSO EM CURITIBA

Curitiba sediará, de 21 a 23 de julho, a segunda edição do Congresso Brasileiro de Metodologias Ágeis de Software.
Considerado o maior encontro do segmento na América Latina, o evento é promovido pelo Centro Internacional de Tecnologia de Software (CTIS), em parceria com a Objetive Solutions - pioneira em Extreme Programing no Brasi, e a Faculdade de Informática e Administração Paulista (FIAP).
Apresentar as principais tendências desta metodologia que vem revolucionando o mercado mundial é a proposta do Congresso, que contará em sua programação com diversos palestrantes nacionais e internacionais.
Informações no endereço www.xispc.com.br/evento2004

LIVRO DETALHA IMPLEMENTAÇÃO, GESTÃO E OPERAÇÃO DE FÁBRICAS DE SOFTWARE

"Fábrica de Software - Implantação e Gestão de Operações" é um livro pioneiro no mercado editorial, não só brasileiro como também mundial, ao apresentar, em detalhes, como projetar, implantar e contratar uma fábrica de software; bem como estruturar uma plataforma de offshore de desenvolvimento.
O livro é uma referência para todos os que queiram entender melhor os aspectos que contornam uma operação de desenvolvimento de software em larga escala. A experiência vivida pelos autores - Descartes de Souza Teixeira, diretor-executivo do Instituto de Tecnologia de Software (ITS) e o professor Aguinaldo Aragon Fernandes, diretor da PRIZM Consulting Group e membro do Conselho Técnico Científico do Instituto - e o conhecimento que trazem sobre tecnologia da informação, representam inestimáveis fontes de informação para toda a comunidade brasileira do setor.



Com 308 páginas e em sua primeira edição, "Fábrica de Software - Implantação e Gestão de Operações" é uma publicação da editora Atlas e pelo preço sugerido de venda de R$ 48,00 torna-se um livro acessível aos que desejam conhecer mais sobre o conceito de fábrica de software.

Confesso que ainda não o li, mas cumprimento os autores pela inciativa pioneira do assunto.



Posted by Felipe Machado em Quinta-feira, Maio 20, 2004

Comentários:

Para os Java Boys

As vezes mais importante do que escrever é mostrar quem escreve, sabe e o faz bem feito

A imagem a seguir é de um execelente blog de Java que localizei por acaso

Vale a pena ler e interagir com este blog, pois tecnicamente é muito bom mesmo, clique na imagem e conheça



Bom proveito

Posted by Felipe Machado em Terça-feira, Maio 18, 2004

Comentários:

Follow, Don't Lead

(COMPUTERWORLD) - One powerful way to reduce costs without forgoing new systems is simply to spend more slowly. The rapid, ongoing fall in IT prices means that even small delays in purchases can dramatically reduce the cost of achieving a given level of IT functionality.

see more in ComputerWord


Posted by Felipe Machado em Segunda-feira, Maio 17, 2004

Comentários:

Offshore Winds: Outsourcing

The anti-offshoring voices are becoming louder, even as the imperative to cut costs by sending jobs out of the country grows. Most experts concede that the trend will not be slowed much by the politics of the U.S. job market¿the benefits are simply too attractive to ignore. But don't underestimate the skills needed to manage the risks involved.

See more in CIO INSIGHT


Posted by Felipe Machado em Domingo, Maio 16, 2004

Comentários: