Re:The real 90s versus outdated 00s software
on
Java Is So 90s
·
· Score: 1
Java is definitely outdated and it's fall was absolutely predictable. Virtually every programming/scripting language developed by the corporate world (with the exception of Cobal) has been short lived. Look at the history dating back to the mid 50s.
For survivors, languages designed to solve specific problems have enjoyed a long life. Fortran was designed by John Backus to facilitate converting formulas to machine language (formula translator), Basic came out of Dartmouth to make teaching programming easy, B came from Ken Thompson at Bell Labs which led to Dennis Richie's C in order to overcome obstacles with the Multics/PL-1 project. More recently, Perl, PHP etc - also were driven by individuals solving specific problems.
What fell? PL/1, designed by IBM to combine the utility of Cobol and Fortran failed. The Defense Department jumped in and designed ADA and so on until we arrive at Java. Designed by Sun - again, not a response to a specific issue but rather a coporate strategy to solve classes of perceived problems.
Bottom line - programming/scripting languages, designed by individuals to solve individual problems seem to enjoy a longer life than languages designed by corporate groups that address
large classes of problems (to increase market share?).
And another point - typically languages initiated by individuals have a shallow learning curve (remember printf("Hello world");) whereas languages initiated by groups tend to be too feature-rich (look at PL/1).
The remarkable lesson for me is how a few individuals addressed their own problems in such a way that the resulting language could be extended while maintaining a very shallow learning curve.
But RedHat is not in Silicon Valley. In Raleigh-Durham , $20M is a lot of money. This investment is an interesting move to opening up more resources for the open source community
I don't get it. We have been using the National Hurricane Service RSS feeds for about 8 months - at no cost of course, and recently the National Weather Service made available a SOAP interface for extracting regional weather data. Here is a quote from their web page:
Why provide an Experimental NDFD XML service?
The National Weather Service is striving to serve societys needs for weather information by evolving its services from a text-based paradigm to one based on making NWS information available quickly, efficiently, and in convenient and understandable forms. The NDFD is one example of this transformation. NDFD XML takes yet another step towards a digital services era by making NDFD data available for computer to computer transfer and processing. NWS customers and partners can then enhance the value of NDFD data through the creation of value added products.
Well you would think so. For 3 months I have been trying to unsubscribe to PC Magazines almost daily emails. I have clicked on unsubscribe until my fingernails fell off. I wrote to abuse@enews.pcmag.com, I wrote to security@enews.pcmag.com - no replies, no nothing except continue endless stream of announcements. Sometimes life is simply harder than it is supposed to be:)
Seems to me that the issue here is not what most people think, but rather what actually is. The quoted dictionary definition is a rather precise definition and the Linux Kernel group of developers surely qualifies as an organization. What is missing from your "most" perspective is that neither Linus nor Linux are "most" in any manner of speaking. Rather, Linus and Linux reflect the leading edge of a paradigm shift in software development - where company boundaries are blurred.
Within this frame of reference, a company is the residual of the past software development paradigm and the organization is a step in a new software development paradigm.
That Torvalds duties differ by miles from that of Ballmer and Gates is a sign of genius - Linus can manage an open source development organization without the traditional management hierarchy that is managed by Ballmer/Gates and all the rest of 'em.
A long long time ago (1970s) Charlie Molnar, designer of the Linc tape (the Linc computer was
an NIH funded (late 1960s) minicomputer that evolved into the PDP 8 and pushed DEC into the minicompuer business) explored asynchronous computing. Along the way they discovered synchronizer failure - i.e. the inability to reliably synchronize asyncronous subsystems - see Chaney, T.J. and Molnar, C.E. 1973. Anomalous behavior of synchronizer and arbiter circuits. IEEE Trans. Comp. pages 421-422. The bottom line is that it is physically impossible to guarantee that the data setup requirements (the minimum time the data must be asserted before it can be reliably clocked into the flip flop) of a flip flop can be met when the clock is asserted by one async component and the data are asserted by another async component. To my knowledge, this fundamental limitation has never been overcome.
It is useful to remember that our government consists of three branches: Executive, Legislative and Judicial. The Judicial Branch, that ruled on the do not call list is not the same bunch that brought us the Patriot Act (Legislative Branch). Sometimes, the checks and balances in Washington even work.
How long have influential members of the community be predicting the demise of tape and tape systems? When the going gets hard, the community seems to alywas find ways to overcome the obstacles predicted by the experts.
In the days before.... I carried both a swiss army knife and leatherman with me - to collaborators working in labs around the world. No problem ever until the fall of 2001. Then my leatherman and swiss army knife were forbidden. Most recently Paris (CGD) "borrowed" my laser pointer as a security threat - after having travelled through security in Russia, India, Thailand, Oman, UAE, Greece, Germany, UK, Belgium. This time, it travelled into France - but did not manage to exit France. When I asked if I could keep it by disabling it and giving them the batteries - then answer was clearly NO.
So take care with the flash swiss army knives - I suspect there is a great marketplace run by airport security groups around the world.
Sun has tried several times to build diskless workstations such that your personal config followed you from box to box. It never took off, in my opinion, because server failure put you out of business. But any of the high capacity solid state stores, CF, memory sticks etc - provide the option for you to carry your config, and critical documents around with you - plug into a true diskless station and pick up where you left off.
This laptop question just might be a good test bed to test this concept.
Ten years ago, I taught at the Indian Institute of Technology - Madras, and met Prof. Kalyana Krishnan. At that time Krishnan was struggling with how to render characters on a web page so that he would produce Hindi, Tamil, etc web pages. Over the years, Krishnan's project has expanded, now has voice rendering of web pages and was recently recognized as a major innovation benefiting many in India. His
project website will give you an idea of the tools he and his students are bringing to all of India. e-Governance is a small segment of the challenges facing India. Skilled practitioners, coming from the IITs in India are effectively penetrating the digital divide.
Fred Brooks told wonderful stories in the
The Mythical Man Month about software development and one of the best was how second versions tended to have all the features that were missing from the first version to the point where the language or system's use was constrained by "too many verbs" (in a Mozart sence). Most 1st versions are nice 80% solutions, lean and mean.
My dream is that a redesigned Internet Protocol will continue to be lean and mean, and not over-bloated with "if we only had this feature then we could do that".
There is an interesting history of programming languages. By and large (with a few exceptions), languages designed by industry to meet perceived needs are proprietary and have, over the years failed. The major exception is Cobol. But where is PL/1, ADDA etc. On the other hand, languages that were designed to solve problems, and typically prototyped by a very small group, have succeeded. Consider Fortran, Basic, C, Perl, PHP and the list goes on. Perceived-needs-driven languages tend to be overly rich (too many verbs) and have steep learning curves whereas problem-driven languages tend to have only the essential verbs, are readily extended with modules and tend have small learning curves (printf "Hello world\n").
If Sun opens the Java source code, there will be forks - but my speculation is that the forks will move toward a less rich language for the masses that is an 80% solution. The 100% solution will require modular extentions while leaving the core untouched. A new standard JAVA-like language will likely arise - there will be back compatibility problems etc - but the end result will likely be a more generic platform independent language tool whose features are derived from the real problems it solves. It will live because it will enable a large number of non-experts to join us in problem solving.
What use are open standards if one can patent a particular work derived from an open standard? For example, imagine patenting an extension of the HTTP protocol and then charging royalties each time it is invoked.
Seems to me that patent law needs to be updated to reflect today's challenges and today's technology. Otherwise, there will be other obvious cracks where folks can patent a derivative or specific implementation of an open work. This is the work for Congress - who are the sympathetic ears in Congress?
Does the notion that a small group, using Internet-centric communication, coordination and managment imply that the days of the political lobbyist are numbered? Does it imply that we can create an organized approach to reaffirming
Fair Use and restructuring the
DMCA. If Howard Dean can use Internet-centricity to overthrow the Democratic Party then surely the Slashdot community can coordinate moves to restructure DMCA along user-centric lines.
Open source excels at the individual level - we are free to do what we want. But for many of us - the issue is also about sustainability. A project with too many forks is less likely to create a critical mass of contributors, necessary to keep the fork alive. A project with too few forks may actually stifle innovation.
Management of the huge projects (apache, linux kernel) have managed forking while keeping innovation alive. Less huge projects (KDE, Gnome) have enough critical mass to be sustainable and more freedom within the desktop forks. Something from my group would die if it forked - die from starvation of ideas and loss of critical mass of ideas and energy.
It is useful explore the tension that exists between Open Source and both SCO and Microsoft within the light of the Declaration of Independence. Down about the 1st sentence of the 2nd paragraph - we read: We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty and the pursuit of Happiness.
It seems to me that the Open Source movement is all about liberty (Stallman's definition of free) and pursuit of happiness and is just the tip of the iceberg that results from Internet connectivity creating such large communities that further enable us to pursue Life, Liberty and Happiness. Ballmer wants "butts on the line" to get productivity. Open Source Development wants committed individuals. The difference is that Ballmer can only draw on the employees of Microsoft whereas OSD draws on the world. Similarly SCO finds GPL as a tool to promote open communication in conflict with the constitution, and implys that open communication attenuates innovation.
It seems to me we are on the leading edge of a paradigm shift in group processes - and software development is leading the way. I find it useful to be reminded about what our political system is supposed to be all about.
While RedHat is terminating support of the free distribution, it appears that Fedora has free support for Fedora Core - same up2date client, same registration process. So, it seems that RedHat is removing free RedHat production and support from its books so that it can make real money - which in turn provides some of the paid manpower that makes free Fedora production and support actually remain available at either no cost or a modest cost.
RIAA has used the argument that file sharing has reduced sales of music CDs over the past several years. If this is correct, then there should be a direct correlation between file deletions and CD sales (assuming folks who listed to their music files will continue to listen to something). It will be interesting to track CD sales and see if there is a turn-around in the record market. If there is no turn-around, it suggests that excessive price is the culprit, not file sharing - a quiet victory for our community.
Johnson states in his article that if you "want to learn gardening tips, or basically want to know anything about flowers that doesn't involve purchasing them online, you have to wade through a sea of florists to find what you're looking for". So I searched on gardening tips with Google. Guess what - the first page of results were all about gardening tips. The only links to florists were on the right side bar - where they belong. Seems to me Google got it right.
In addition to solving computationally complex problems involving pattern matching (DNA computing), the ciruitry in the eye can do some
real time processing. The retina is wired in a way that edges are automatically computed. It works like this - the retina is basically an array of light detectors (rods and cones) connected to nerve cells.The rod and cone generate small electrical signals when they receive a photon (a single quantum) of light. This signal flows into a the central region of a cell called a horizontal cell. The axon of the horizontal gathers signals from nearby cells, but of the opposite polarity and the amount of signal degradation is proportional to the distance from the center of the cell. Thus the cell computes a weighted average - for two neighbors - the weights might be -1/2 -1 3 -1 -1/2 - the result is something that approximates a laplacian, a 2nd derivative in space. For a uniform illuminated field, the 2nd derivative is zero - but when there is an abrupt change in illumination, the derivative is non-zero. If there is motion, then the array of horizontal cells will produce a signal that spatially follows the edge. Neat stuff
Slashdot Mirror
For survivors, languages designed to solve specific problems have enjoyed a long life. Fortran was designed by John Backus to facilitate converting formulas to machine language (formula translator), Basic came out of Dartmouth to make teaching programming easy, B came from Ken Thompson at Bell Labs which led to Dennis Richie's C in order to overcome obstacles with the Multics/PL-1 project. More recently, Perl, PHP etc - also were driven by individuals solving specific problems.
What fell? PL/1, designed by IBM to combine the utility of Cobol and Fortran failed. The Defense Department jumped in and designed ADA and so on until we arrive at Java. Designed by Sun - again, not a response to a specific issue but rather a coporate strategy to solve classes of perceived problems.
Bottom line - programming/scripting languages, designed by individuals to solve individual problems seem to enjoy a longer life than languages designed by corporate groups that address large classes of problems (to increase market share?).
And another point - typically languages initiated by individuals have a shallow learning curve (remember printf("Hello world");) whereas languages initiated by groups tend to be too feature-rich (look at PL/1). The remarkable lesson for me is how a few individuals addressed their own problems in such a way that the resulting language could be extended while maintaining a very shallow learning curve.
But RedHat is not in Silicon Valley. In Raleigh-Durham , $20M is a lot of money. This investment is an interesting move to opening up more resources for the open source community
Well you would think so. For 3 months I have been trying to unsubscribe to PC Magazines almost daily emails. I have clicked on unsubscribe until my fingernails fell off. I wrote to abuse@enews.pcmag.com, I wrote to security@enews.pcmag.com - no replies, no nothing except continue endless stream of announcements. Sometimes life is simply harder than it is supposed to be :)
Personally I prefer REAL intelligence of Google to the ARTIFICIAL intelligence of the Chinese masters.
Within this frame of reference, a company is the residual of the past software development paradigm and the organization is a step in a new software development paradigm.
That Torvalds duties differ by miles from that of Ballmer and Gates is a sign of genius - Linus can manage an open source development organization without the traditional management hierarchy that is managed by Ballmer/Gates and all the rest of 'em.
A long long time ago (1970s) Charlie Molnar, designer of the Linc tape (the Linc computer was an NIH funded (late 1960s) minicomputer that evolved into the PDP 8 and pushed DEC into the minicompuer business) explored asynchronous computing. Along the way they discovered synchronizer failure - i.e. the inability to reliably synchronize asyncronous subsystems - see Chaney, T.J. and Molnar, C.E. 1973. Anomalous behavior of synchronizer and arbiter circuits. IEEE Trans. Comp. pages 421-422. The bottom line is that it is physically impossible to guarantee that the data setup requirements (the minimum time the data must be asserted before it can be reliably clocked into the flip flop) of a flip flop can be met when the clock is asserted by one async component and the data are asserted by another async component. To my knowledge, this fundamental limitation has never been overcome.
It is useful to remember that our government consists of three branches: Executive, Legislative and Judicial. The Judicial Branch, that ruled on the do not call list is not the same bunch that brought us the Patriot Act (Legislative Branch). Sometimes, the checks and balances in Washington even work.
How long have influential members of the community be predicting the demise of tape and tape systems? When the going gets hard, the community seems to alywas find ways to overcome the obstacles predicted by the experts.
So take care with the flash swiss army knives - I suspect there is a great marketplace run by airport security groups around the world.
This laptop question just might be a good test bed to test this concept.
Ten years ago, I taught at the Indian Institute of Technology - Madras, and met Prof. Kalyana Krishnan. At that time Krishnan was struggling with how to render characters on a web page so that he would produce Hindi, Tamil, etc web pages. Over the years, Krishnan's project has expanded, now has voice rendering of web pages and was recently recognized as a major innovation benefiting many in India. His project website will give you an idea of the tools he and his students are bringing to all of India. e-Governance is a small segment of the challenges facing India. Skilled practitioners, coming from the IITs in India are effectively penetrating the digital divide.
My dream is that a redesigned Internet Protocol will continue to be lean and mean, and not over-bloated with "if we only had this feature then we could do that".
If Sun opens the Java source code, there will be forks - but my speculation is that the forks will move toward a less rich language for the masses that is an 80% solution. The 100% solution will require modular extentions while leaving the core untouched. A new standard JAVA-like language will likely arise - there will be back compatibility problems etc - but the end result will likely be a more generic platform independent language tool whose features are derived from the real problems it solves. It will live because it will enable a large number of non-experts to join us in problem solving.
Seems to me that patent law needs to be updated to reflect today's challenges and today's technology. Otherwise, there will be other obvious cracks where folks can patent a derivative or specific implementation of an open work. This is the work for Congress - who are the sympathetic ears in Congress?
Does the notion that a small group, using Internet-centric communication, coordination and managment imply that the days of the political lobbyist are numbered? Does it imply that we can create an organized approach to reaffirming Fair Use and restructuring the DMCA. If Howard Dean can use Internet-centricity to overthrow the Democratic Party then surely the Slashdot community can coordinate moves to restructure DMCA along user-centric lines.
Open source excels at the individual level - we are free to do what we want. But for many of us - the issue is also about sustainability. A project with too many forks is less likely to create a critical mass of contributors, necessary to keep the fork alive. A project with too few forks may actually stifle innovation. Management of the huge projects (apache, linux kernel) have managed forking while keeping innovation alive. Less huge projects (KDE, Gnome) have enough critical mass to be sustainable and more freedom within the desktop forks. Something from my group would die if it forked - die from starvation of ideas and loss of critical mass of ideas and energy.
It seems to me that the Open Source movement is all about liberty (Stallman's definition of free) and pursuit of happiness and is just the tip of the iceberg that results from Internet connectivity creating such large communities that further enable us to pursue Life, Liberty and Happiness. Ballmer wants "butts on the line" to get productivity. Open Source Development wants committed individuals. The difference is that Ballmer can only draw on the employees of Microsoft whereas OSD draws on the world. Similarly SCO finds GPL as a tool to promote open communication in conflict with the constitution, and implys that open communication attenuates innovation.
It seems to me we are on the leading edge of a paradigm shift in group processes - and software development is leading the way. I find it useful to be reminded about what our political system is supposed to be all about.
While RedHat is terminating support of the free distribution, it appears that Fedora has free support for Fedora Core - same up2date client, same registration process. So, it seems that RedHat is removing free RedHat production and support from its books so that it can make real money - which in turn provides some of the paid manpower that makes free Fedora production and support actually remain available at either no cost or a modest cost.
RIAA has used the argument that file sharing has reduced sales of music CDs over the past several years. If this is correct, then there should be a direct correlation between file deletions and CD sales (assuming folks who listed to their music files will continue to listen to something). It will be interesting to track CD sales and see if there is a turn-around in the record market. If there is no turn-around, it suggests that excessive price is the culprit, not file sharing - a quiet victory for our community.
Johnson states in his article that if you "want to learn gardening tips, or basically want to know anything about flowers that doesn't involve purchasing them online, you have to wade through a sea of florists to find what you're looking for". So I searched on gardening tips with Google. Guess what - the first page of results were all about gardening tips. The only links to florists were on the right side bar - where they belong. Seems to me Google got it right.
In addition to solving computationally complex problems involving pattern matching (DNA computing), the ciruitry in the eye can do some real time processing. The retina is wired in a way that edges are automatically computed. It works like this - the retina is basically an array of light detectors (rods and cones) connected to nerve cells.The rod and cone generate small electrical signals when they receive a photon (a single quantum) of light. This signal flows into a the central region of a cell called a horizontal cell. The axon of the horizontal gathers signals from nearby cells, but of the opposite polarity and the amount of signal degradation is proportional to the distance from the center of the cell. Thus the cell computes a weighted average - for two neighbors - the weights might be -1/2 -1 3 -1 -1/2 - the result is something that approximates a laplacian, a 2nd derivative in space. For a uniform illuminated field, the 2nd derivative is zero - but when there is an abrupt change in illumination, the derivative is non-zero. If there is motion, then the array of horizontal cells will produce a signal that spatially follows the edge. Neat stuff