I think I even registered a slackware install way back in '96. I had no idea the linux counter still was alive. In any case I think it safe to say that most people who installed linux over the past few years are not aware of the existence of the good old counter so it must be way of by any standards and removing 90000 unverified users won't do much good.
DSL works fine in the Netherlands. The problem in the US is badly managed telecom companies trying to revive their business using a silver bullet called DSL.
In the Netherlands the copper network is in good shape and the largest problem has been getting the local telecom switches converted (a process that is still not completed everywhere). In most of the larger cities people have a choice between cable and DSL. DSL tends to be bit more reliable but also more expensive and cable has a bad reputation mainly due to the fact that companies like @home are active on the isp site there. The competition between cable and DSL has stimulated quality improvements in both.
I've had my DSL connection for nearly a year now. Apart from some technical problems in the beginning, I've enjoyed a good connection and get exactly what I payed for. In any case, DSL and cable are of course a temporary solution until we all can have a fiber optic connection.
Of course in Europe, local telephone connections not for free (like in the US), so people are more likely to take DSL to save money. Basically if, like me, you want to be online a lot, DSL is much cheaper than a regular modem connection. In the US your local connection is for free so you can be online all day relatively cheaply.
You'd still be using the Java API that way so it wouldn't help you much. The whole point of migrating to.Net is that you end up using.Net. Transforming the syntax doesn't really help since both languages are more or less equal in capabilities.
MS knew exactly what they were doing when they broke compatibility (presumably they can afford the legal staff to read a license) and it had nothing to do with improving the product. They expected that they could get away with it but were proven wrong.
They did implement AWT BTW. The problem was the way they called native code (i.e. they didn't use JNI). The only reason they did not support JNI was because that would allow them to create platform specific APIs in addition to platform specific implementation. MS always had the option to implement the necessary APIs but they refused to do so. Consequently, MS is entrirely to blame for the death of J++.
J# is Java without the Java platform (vm + API), except for the obsolete, partial jdk1.1.4 API. It is generally agreed that what makes Java so powerfull is the vm + API. The language itself is not very exciting really (your average OO language).
Re:Get thee to the C# decompiler.
on
J#
·
· Score: 3
Decompilers are close to useless for maintenance activities unless you don't have the sourcecode (so either you threw it away, duh, or you're not supposed to have access to it). You do maintenance on sourcecode, preferably properly commented source code, preferably source code you are familiar with.
Whether you compile to.Net or to bytecode is irrelevant. There's really only one certainty: migrating to.Net will generate additional maintenance activities since you will have to test whether it works and almost certainly you will encounter small problems. Once you have migrated and supposedly actually use the binaries you will find that you need to fix little things and even add new features from time to time. So you will keep tinkering with your old J++ code.
It is worthwhile to mention that J# is not really a migration tool since there is no conversion of source code taking place into one of the typical.Net languages. Also J# apps still use the rather limited 1.1.4 API and not the.Net API so it is really a tool to communicate with legacy J++ applications rather than to migrate those legacy applications (which eventually you may need to do after all). And once more: the whole problem of J++ applications being legacy software is because MS decided to abandon J++. First they charge you for a J++ development environment and now they charge you again for some crappy solution to let you keep using your J++ stuff. Probably the reason you decided to use J++ in the first place was to be compatible with the other wonderful windows stuff you paid for. You were screwed three times! If I had used J++ in the past, I'd be very pissed off. This also shows that you have to be careful to invest too much into C# or Visual Basic or.Net because MS may suddenly decide that they need some extra revenue and change things so much that what you invested in heavily suddenly becomes legacy software. Visual Basic changed substantially during its existence and generally 'migrating' apps between different versions cannot be done fully automatically (i.e. it will cost you). Just like with office it is highly debatable if the changes are worthwhile and that breaking compatibility is really necessary but you have no alternative so you upgrade and revise your old stuff.
jdk1.0.2 code still compiles on jdk1.4 (ok, minus some minor API changes which were necessary and are generally easy to fix). The resulting bytecode can still be executed on a jdk1.0.2 VM. Try to do that with VB applications written in 1995 (when jdk1.0.2 was released) on the latest version of VB and you will see my point.
increasing size of sun VM?????
on
J#
·
· Score: 5, Insightful
I keep reading about this on sites like this. It should be pointed out that even the JRE (java run-time environment) for jdk 1.4 is well below 10 megabytes (mainly depending on the platform). Of course if you download the full jdk you have a bigger download, mainly because that includes, among others, various tools and the source code for most of the API. But even then we are talking about 30 or 40 MB.
Go check out Opera or netscape, both have an optional download for the JRE 1.3.x. I think it was about 6MB. The JRE includes everything you need to run Java applications. It is hardly bigger than MS jvm and does a lot more.
Incidently, there is currently a beta of an enhanced 1.3.1 JDK that includes an activex component that fully replaces microsoft's JVM. Yes that's right, you can now run all your applets in IE using jre 1.3.1. Of course it doesn't support the MS specific extensions of the JVM.
Re:Of course "no one" will use it
on
J#
·
· Score: 5, Interesting
Even as a migration tool its use is limited to projects using the partial jdk1.1.4 API MS supports. Basically any serious Java project nowadays is written using the Java 2 (i.e. 1.2 and upwards) API. And if they are written to the 1.1 API they likely use jdk1.1.8 rather than 1.1.4.
So basically it is a migration tool for J++ applications. Considering that is a MS product, it makes you wonder if this is the best MS can pull of after all the sole reason migration is needed is because MS decided to drop Java support (so they already screwed you once).
And even for J++ it is limited since it only allows you to compile source code, you lose information stored in e.g. forms, project files and so on. In addition, Java objects tend to be closesely tied to the Java API and reusing them basically means you are using the Java. So you might as well go for the real thing (including richer API, better performance and so on) and use the com bridge to communicate with the objects.
In any case, if this is just a migration tool, MS is going through a hell of a lot of trouble to present it as a Java alternative.
Some advise to people considering to use it:
- After 'migration' it is still Java code you are using. It won't be much faster and you will still have to maintain it.
- MS is not very Java friendly, they might drop support for the migration tool at their convenience. The long term strategy is C# (if it ever takes of), not J#.
- Migration to real Java is probably much easier unless you heavily rely on MS specific APIs
- There are ways of letting Java objects talk to COM objects (and consequently also.Net objects) that don't require recompilation.
The whole problem is a simple economic problem. In the old days communicating peer reviewed papers to colleagues involved getting a printed copy to them. It was not more than logical to use the services of publishers to take care of the logistics and naturally they were in it for the money. In addition the publisher needed copyright in order to be allowed to print.
Nowadays there is an alternative means of getting the scientific content distributed: the internet. Many scientists have already discovered the internet as a means of distributing and obtaining papers. I can't remember the last time I actually had to go to a library to look up a paper. The internet totally eliminates the need for a publisher and could potentially greatly reduce the cost involved with publishing articles without affecting the level of quality of peer review in any way. In addition the internet provides additional possibilities in the form of for example moderated forums, mailinglists etc.
Unfortunately the publishers have built up an enormous power and are very reluctant to release it. Scientific journals are very profitable since they require a minimum of investment on their site and they have a small but loyal audience who are willing to pay relatively large amounts for journals. However over the past few years there has been a pressure to move to an internet based distribution of papers and journals. Despite this publishers still receive enormous amounts of subscription fees. At our university department, the library sucks up obscene amounts of our budget for online subscriptions and regular subscriptions. All this money (mostly coming from our government) flows to the publishers who add little or no value to the scientific product since they completely rely on external input (voluntary peer review, editing and academic authorship).
Most journals of significance in my field are either associated with ACM or IEEE, both of which are non profit organizations, both of which are charging significant amounts of money for membership and access to papers. A step forward would be if these two organization would abandon the publishing process and focus on internet publishing only. If those two organizations set the necessary steps, the rest would have to follow. As an author who has contributed to several IEEE conferences and who has even been a IEEE member (I have resigned because of the outrageous lack of value added by a membership) I would very much like to see this happen. After all they are supposed to represent our interests rather than work against us.
Basically I would like to see two things:
- Authors should retain copyright over their articles. The arrangement proposed for the ML journal seems reasonable, perhaps some GNU like legal document could be created to formalize it.
- Papers should be available and searchable at no cost. Currently I have to pay to be able to be able to download my own articles from the IEEE. I find that offensive (luckily I have copies on my homepage:-).
In addition I would also like to see some innovations:
- XML based structure for papers to enhance searching.
- Addition of the before mentioned forums.
- distributed storage of content ala freenet.
If these requirements are met (even minus the innovations), I'll gladly renew my IEEE subscription and sign up for the ACM as well. I'm fully confident that with the logistics of the physical publishing process eliminated the membership fees can be reduced significantly. Alternatively this money can be used to set up more conferences and finance promising researchers.
However, currently I feel that neither of these organizations are working in my interests.
By definition, infrastructure should encapsulate every feature that is common to the applications that use it. The problem with UNIX is that such infrastructure is lacking. This has lead to a situation where in order to be able to compete with ms windows (which does have such infrastructure), applications have to include everything and the kitchen sink. Look at star office, it comes with its own widgets, its own printing subsystem, its own way of embedding components and until the recent beta it even came with its own desktop. Sure it is integrated with Gnome/KDE to some extent but since it also needs to be portable across the two it contains a lot of duplicated functionality.
KDE developers have understood this and are currently working to deliver such an infrastructure. Ignorant critics complain about konquerors ability to be both a browser and a file manager. However, once you understand what infrastructure is supposed to be you recognize that same ability as a good working infrastructure. File managers and browsers have a lot in common and therefore you might as well integrate them so that you don't have to invent the light twice.
The UNIX philosophy is to make something small and only once. Most unix applications only meet the first part of that philosophy and have to duplicate everything and the kitchensink because it is either not present in the infrastructure or not consistent enough to allow for easy integration (this should also be facilitated by the infrastructure).
I simply use visio to do drawings in framemaker. Works fairly well. I actually like framemaker a lot. It's a complex application with a lot of features which at the very least are non-trivial. However, once you get into it it seems like the people who built it anticipated what you really need.
Object linking and embedding supports both types of links you are talking about (even though it defaults to embed rather than link). On windows star office just uses the windows infrastructure so it is not different from ms office in that respect.
I have been reading very positive reviews about SO during the past few days. So, I decided to give it a try myself. Here's an overview of my experiences.
Let me start by saying that it looks promising, despite some obvious flaws which I will detail below.
Installation
I installed it on windows 2000. The installer looks pretty and userfriendly and seems to be doing what it is supposed to be doing in a straightforward way. I haven't tried the uninstall yet (duh) but I trust that will be equally good.
Compatibility with office
I normally use framemaker to write my papers but occasionally I have to deliver word documents. Star office certainly supports word better than framemaker. However, I found that it had some trouble with the word documents I had. All of these documents are rather complex and they use tables, crossreferences, images and paragraph styles. I was impressed that the document was successfully imported. However, there were a few errors that though easy to correct would have lead to visible errors if I had just printed them without looking at them. I had a similar experience with my powerpoint documents. Near perfect but not entirely perfect. Conclusion: you will need to revise imported non trivial office documents to make sure everything is the way as intended.
User interface
From the point of view of features most of it seems to be there and functioning. However, it is far from pretty. One of the key features of office is that it offers a polished userinterface. Obviously MS has a few graphical designers who know what they are doing. I can't say the same about SO it all looks rather primitive when compared to office (far better than framemaker though:-).
In addition to the look and feel I found some of the dialogs a bit confusing. Some buttons don't have tooltips and there are slight inconsistencies, especially in the more advanced dialogs.
Features
After having played with star writer a bit I would say that I would actually prefer it over word and may even consider to use it as a replacement for framemaker in the future. As mentioned before, I use stuff like crossreferences and paragraph styles quite heavily. The prime reason I am using framemaker rather than word is because word is designed for stupid users and tends to "improve" the deisgn of your document on the fly which is a PITA if you do know what you are doing. Aside from that it's handling of graphics is really bad and you are at a constant risk of seriously messing up the layout of your document. Star office doesn't seem to suffer from these flaws and in addition has a build in literature references database!!!! That last feature (once I figure out how to use it because the GUI is a mess) might pull me over. Also I like the fact that everything is stored as XML and I am looking forward to any add on utilities that rely on this.
So in conclusion, star office looks very promising. I have only looked at star writer and compared it to word 2000. Word 2000 arguably is more polished and user friendly. In terms of features the two suits can compete even though star writer is lacking some of the popular word features (e.g. grammar checking). However, it also has features that either work better than or are not even present in word (e.g. the bibliography tool).
It looks like I am going to give it a try but it also looks like I won't be recommending as an alternative to office to less advanced users. If you know what you are doing and are equally annoyed with word as I am you might actually like it (otherwise I recommend you take a look at framemaker). It is not the office killer it is advertised to be but it certainly looks like you could use it for most of the stuff you would MS office for, especially when handling more complex documents. It's biggest problem is the user interface which just doesn't have the same quality as MS office.
Encryption is rather different than a gun in a few respects:
1) - the tools are software: duplication is easy. Guns are hardware and sophisticated knowledge is needed to make them.
2) - the algorithms are well known: you can make your own tools (without the backdoors). Building your own guns is a bit harder (though not impossible)
3) - there are open source tools (you don't even have to go through step 2 to obtain tools free from backdoors). Although the US occasionally hands out guns (e.g. stinger missiles to the afghan resistance a.k.a. taliban in the eighties), in general selling arms is profitable business.
Now about guns: you need a gun + an idiot to pull the trigger to kill people. Both prerequisites are available in large quantities in the US. In western europe, guns are a bit harder to get so we have less casualties as the result of guns (check the statistics if you like). Obviously, removing guns from society helps reduce the amount of people dying from guns. Doing so is a problem in the US however since billions of guns have been sold there in the recent centuries. So if you are in the US you are fucked, people around you are nuts and have easy access to guns. One day your nice neighbour or colleague may have a bad day and pull his guns on you (which he can buy legally and keep in his house).
Now lets turn to the real issue: why is the US pushing backdoors in encryption software: industrial espionage. Being able to tap in on information banks and businesses exchange throughout the world is very profitable business. A terrorist will just use illegal/free tools (probably on a illegal version of win XP or whatever). If there's one thing you can be sure of: terrorists don't like the US and they are not bloody likely to stimulate the US economy by actually paying for software produced in the US. What do you think? Bin Laden will actually log on to MSN and chat with his colleagues??? Come on!
The US government is using this situation to rearrange the world to make it a little bit more comfortable for the US leaders. Aguably the WTC tragedy was the best thing to happen to them in years. Some impopular anti-terrorist/anti-human rights laws can be pushed through. Suddenly they can be friends with Pakistan (a few weeks ago still referred to as a rogue state that we should be protected from by a missile shield). Everybody turns a blind eye while they whipe the Taliban of the earth and even Khatami is suddenly being friendly on behalf of Iran. In addition some former Soviet republics who happen to play an important role in producing and transporting oil are also the US' best friends.
It is touching to see all this friendship bloom. Unfortunately it is at the cost of millions of innocent Afghan civilians, already in big trouble because of the previous civil wars. What happened to New York was bad but the opportunistic way the US government is dealing with the situation is sickening.
What MS did is remove the ms vm and disable netscape style plugins. The old sun plugin was a netscape style pplugin.
Sun now has a beta version of jdk1.3.1 (in their early access program) that replaces the MS vm entirely. It is an activex plugin and it will even run applets defined in applet tags.
If you go to www.javasoft.com/jdc and sign in (for free) you can download it. You'll notice that sun has put up a call to testers to test it with existing applets in order to find compatibility problems.
It works great with IE 6 and I haven't found any applets yet that I can't run. It's stable too, no crashes so far.
"and just doesn't fit in with all the other UI apps I use (it won't follow themes or UI conventions). "
It's funny that you should mention this in a discussion on gui applications for linux. If there's such a thing as a standard look and feel for linux I have yet to encounter it. There's several desktop environments each of which come with their own widget sets, their own way of theming them, their own component models (if any at all) and their own look and feel. Generally you need all of them in order to run common desktop applications. There's no way you can target all those environments as a programmer. And applications written for one environment integrate extremely poor with the other environments (beyond the point of being able to display the user interface).
With Java you want to abstract away from it all so that it works on all platforms. That means you can't rely on native things to work consistently everywhere.
Limewire has achieved that. It's a simple, elegantly designed UI that works the same on each platform. Most of the native competitors pale in comparison and look clumsy when compared. It being crossplatform is vital since gnutella works better if there are more hosts that share files. The limewire people just have to design the GUI once and can focus on adding new features (which they do).
Admittedly there's a problem with integration with the native platform. However, on linux it is absolutely unclear what exactly this native platform is. Should sun integrate the JDK with Gnome, with KDE, with motif with X? Should they create separate jdk's for each environment? What about versions of each environment? The problem is that there is no standard and consequently all sun can do is target the most common denominator. They don't have that problem on mac os X or win32. The JDKs on these platforms generally integrate much nicer. They use file dialogs, the printing facilities, the native 3d, 2d and multimedia libraries, the clipboard and so on. Achieving the same on linux is nearly impossible since there are multiple implementations of each of those components. However, that is a linux problem and not a java problem. IMHO this is the primary reason that linux on the desktop is still not happening outside the developer community. Also I am very pessimistic about these issues being addressed in the near future.
You are of course right. The system is not democratic. Particularly the way of counting the votes and the procedure (or lack thereof) when a conflict arises is contributing to this. The only democratic outcome of the previous election would have been a reelection. Now the outcome of the election was determined by a puppet of the brother of the guy who won.
Yes it is well on its way to become a policestate. Now under the rule of an arguably undemocratically elected president (at least I would have quite a bit of trouble explaining the outcome to the ancient greeks who invented democracy), America is taking one step after another against its own people, who incidentally are cheering (talk about misguided people).
If all this would be limited to just the US I couldn't care less but unfortunately it isn't and europe is pretty much following the US in everything it does.
The recent events, however tragic, make it painfully clear what is wrong with the US. Within minutes after the crash the media machinery puts the video images in an infinite loop feeding them to their public, of course commercial breaks are inserted at regular intervals and CNN is likely to make significant profits in the next few weeks. Soon after, the US president, after being lost for a few hours in some distant US state, appears to make what can only be explained as a christian fundamentalist statement by calling for a prayer. The worst part is that this statement is fueled by opportunism (gotto keep the christian right wing people happy otherwise gore takes over in a few years) rather than true belief.
Now further opportunism dictates to nuke those damn arabs. Never mind the millions of refugees, never mind that the amount of innocent people who will die as a direct consequence of the media show the US army is about to give in the middle east will vastly outnumber the poor souls who lost their lives in new york (btw. a substantial amount of those people were muslim). And never mind that the short attention span of the US media and its audience is never going to outlive the chain of events about to be triggered in the middle east.
The days that reporters reported wars like in Vietnam are long gone. CNN is already being banned from Afganistan and soon the US military can (and will) do whatever pleases them. The gulf war wasn't as clean as CNN wanted us to believe and nor will this war (a b52 is not what I would call an anti terrorist weapon).
Incidently, I recall that a few months ago during the election campaign dubya had some trouble answering the question who the president of pakistan was. I hope his knowledge of this area has improved somewhat by now.
For the time being, enjoy the show just be aware that reality isn't what they show you on TV.
it's not the same thing, trust me on this. The google toolbar is very cool:
type a keyword and
- search the web
- search the site you are looking at
- search the google web directory
- search the archived news groups
- highlight occurences of the terms in the page you are looking at
In addition:
- a google button is present with configuration and links to google, advanced search and so on.
- a ranking is given for the page you are looking at
- a convenient up button is present that moves you one directory up (it's actually a dropdownmenu too so you can select any directory in the path)
- there's a button which shows you the directory in google web directory the site you are watching is indexed in (very handy for finding related material)
- and finally there's an information button which hides useful features as automatic translation, a link to google's cached copy of the page, similar pages and links to pages linking to the page you are watching
To the best of my knowledge opera only offers the first feature (searching the web). After bookmarks, the google toolbar is the single best productivity feature in my browser. It's a real time saver and it unleashes features that you'd otherwise never use because it's too much work. Often google's site search produces better results than the local site's own search option (usually some dumb altavista like engine).
Of course you are right. I was merely claiming that 0 bugs means nothing and is more a statement of testing incompetence than a statement of programming quality. Software with 0 bugs does not exist and good programmers are aware of that.
Of course good design, good methodology and so on helps improve product quality. Bugs found per kloc is certainly a good measure. However you should consider that something like mozilla is measured in MLOC (million lines of code). By your own measures there would be about 2000 bugs in that code. Given that the complexity rises exponentially rather than linearly if the size of a project increases that is probably a very optimistic estimate. I.e. even if all programmers are excellent (very unlikely), mozilla should be full of bugs waiting to be found.
To the credit of Mozilla they manage to track all bugs that are found in a very large development team.
Mozilla's attitude is much more mature than the attitude of e.g. Microsoft (deny until exposed in public) and ultimately leads to better products.
Unfortunately there are very few of these god-like programmers around. Reality dictates that you need a safetynet in the form of a good testing procedure to ensure product quality.
Bugfree software does not exist so the more bugs you find the better your tests are and since tests are also written & performed by programmers this is also a good indication of programmer quality.
A test that would result in 0 bugs found is worthless because that just means that the persons performing the tests are clueless. You don't test to prove that your program has no bugs instead you test to find the most visible bugs. In this respect, testing is very similar to research since you generally should try to disprove a hypothesis (and by failing to do so conclude that it must be valid).
Well, my complaint isn't that there aren't innovative projects (because there are) but that many self proclaimed OSS advocates are mainly concerned with reinventing the wheel and actually have a very conservative attitude towards anything new.
Just to provide some opposition:
ReiserFS. Check out Oracle's and MS plans to replace filesystems with databases. That is innovation, what Reiser does is just an admirable effort to duplicate the journaling feature invented by others.
Berlin. This is a nice project, however, apple has a vector based (pdf) backend for their GUI in mac os X. That is what I call innovative.
Perl, Python, Ruby, TCL, Java (the language, not the VM) are all variants of the same language constructs. Lots of syntactic sugar, good libraries. A lot of hard work went in them. But take a look at intentional programming from microsoft research or multidimensional separation of concerns tools from IBM and you will see my point because that is true innovation.
Some of the stuff I mentioned is actually open source BTW but I'm not discussing the license but the process and the community instead.
Both these projects I would refer to as rather marginal efforts (interesting though). Squeak is essentially a smalltalk clone. I haven't heard of jazz yet (will check it out soon).
I have heard of Alan Kaye. In fact I was about to listen to a keynote by him only last week(unfortunately it was cancelled at the last moment).
It's just C compiled to bytecode. It doesn't have most of the features that made Java popular (e.g. security, dynamicity, reflectiveness because inherent design decisions in the way C programs work inhibit this). In other words it is no replacement.
Don't understand me wrong, maybe integrating such a thing with e.g. the gnu compiler would be useful. It would be great to create a single set of binaries and distribute them to the hubndreds of unix versions and other operating systems. But it has to be clear that this thing is orthogonal to java rather than the same.
In any case, I'm not impressed and even bored. Java is a relatively simple language. Most of the language concepts are easy to grasp, even for novice programmers. Yet people keep comparing it to more primitive languages such as C and suggesting alternatives which are basically C++ improved rather than Java improved. After years of being confronted with a rapidly growing java community, even some C++ programmers begin to appreciate things like garbage collection (after having dismissed it for years based on performance concerns). However, beyond garbage collection they are still largely missing the point (i.e. C is not so cool after all).
What I would like to see from the open source community (perhaps as a proof of concept) is a more ambitious effort, not just a (partial) duplication of features inspired by ignorance rather than innovation. Java has room for lots of improvement, lets set it as a baseline rather than a design goal. Duplicating all of Java's useful features should be a minimum requirement and not the ultimate design goal.
This post is rather harsh, I know. However, I think this is a fundamental problem with open source in general and linux in particular: it's mass production of commodity software components (kernels, compilers, IDEs, word processors), not creation of new ones. If cutting edge technology is your thing, the propietary world is still the source of it (speech recognition, cool 3d technology, AI related improvements to user interfaces, compiler technology, languages,...). It is very rare to see an open source project that does not just duplicate features but instead introduces radically new features and paradigms. There are some research projects that use open source to distribute their stuff but these generally play only a marginal role in the open source community. The big open source projects are all about duplicating and imitating the bigger/better (in most cases) propietary counterparts.
In short I find the open source community boring & backwards. Its contributions are very useful and I use them on a daily basis. However, as a researcher I don't want to wait for thirty years to see things like OO being grudgingly accepted; I hate it when linux is advocated as a windows alternative when the people doing so largely fail to even realize why windows got popular in the first place. The same applies to Java and Visual Basic (yes, from the *evil empire* you understand me correctly). The features that made those things popular have yet to be duplicated. Effords like the one advocated in this article are just so backwards they only make me angry. Go do something useful! Don't waste your time figuring out how to reinvent the wheel, I already have half a dozen.
How about a killer app? The problem is that right now if you start using ipv6 you are pretty much alone. Actually you might as well unplug your network cable, since you won't be able to do much useful stuff with it.
What is needed is ipv6 only services (e.g. mp3 peer2peer filesharing) AND an easy way to get an ipv6 number for your clients/servers that can coexist with your current ipv4 number (i.e. your computer has both an ipv4 and ipv6 number). The easy part is essential because that prevents that people start creating ipv4 gateways to such services (thus removing the need for getting an ipv6 number). There are plenty of ipv6 numbers available so getting and registering one should be made as easy as possible (something like a distributed, global dhcp server that would automatically get you one based on your mac address would come in handy). Come to think of it, why not just automatically convert those mac addresses into ipv6 numbers (mac addresses are supposed to be unique anyway but I'm not entirely sure this is a great idea)
As I understand it, ipv6 can be tunneled over existing ipv4 networks, so it shouldn't be a problem if some routers inbetween ipv6 hosts are ipv4 only.
This would cause the amount of client pc's with ipv6 numbers to gradually grow. Also since lots of PCs don't have static ipv4 numbers, the amount of servers on ipv6 would also grow. Eventually, there will be a critical mass of ipv6 servers and clients and the switch can be made.
Currently there are a lot of p2p applications in development. I imagine, implementing such stuff would be a lot easier using ipv6 with its improved features. Another killerapp could be streaming multimedia (you want to see this great movie, get yourself an ipv6 number now!!).
That has never happened to me and I visit the CNN site on a daily basis. I distrust CNN for another reason: they are a little to friendly with the US government and often act as a PR machine for the US military.
CPU's and memory are cheap, good programmers are expensive. Your assumption that speed is essential is correct but it applies to time to market more than to performance. Execution speed is not so relevant since hardware is relatively cheap compared to programmers (even lousy ones). The year salary of one programmer will likely solve any performance problems you have when converted to extra hardware.
Swing is not lightweight but given a decent PC it will run. In fact all development tools I use are written in Java and I'm not complaining about performance. I'm sorry I can't show you a java program that is faster than the C++ version. However most java applications I could show today, you would still be in development if they were written in C++. That's why java is picking up, you get things done in it.
That's also why Java became popular on the server first. Typically server software is "one of a kind" meaning that each month spent on developing the app adds to the cost. If you have a small development team of say 5 persons costing about 5K $ (good luck finding them that cheap) and you can save 1 month that translates into a lot of hardware. Java development is generally seen as somewhat faster than C++ development so that translates into huge savings on one of a kind software development. Also Java is very scalable, it literally can run on anything from a credit card to a mainframe, so if you have a performance problem, throwing more hardware at it will likely make it go away. And since you develop faster you have something extra to spend. That's why many people creating server software have adopted Java. It works, it works reasonably fast and it scales nicely.
I read the specs. 20 years ago it would have been impressive, but now...
Lets summarize:
- no generics, yet (he was 'thinking' about it)
- I haven't spotted an interface construct like in Java, very useful and no performance penalty -> include it!!
- no dynamic classloading, another useful Java feature
- no reflective features, another useful Java feature.
- no new features worth mentioning
I would propose this language were called Java--. It removes features the author of the spec deems irrelevant and doesn't add any new ones. Java has its flaws but these flaws are tolerated because it also adds useful features.
Incidently, the VM approach has been adopted by MS now in.Net for a very good reason: it adds a great deal of flexibility. Java has already shown that VMs can efficiently execute numerical code. HP has shown that interpreting bytecode can actually be faster than compiling it. There is no sound reason to dismiss the concept of a VM anymore. Java vms are even available on embedded machines now so size is not an issue either.
If the author of the spec still wants to go ahead with implementing D (aka Java--), I'd strongly suggest to make it compile to.Net, the JVM or some other vm (unix currently lacks one) so that you can at least take advantage of the features and libraries offered by such environments.
And finally, there is no market for this language. C/C++ programmers are generally so much in love with the language that they are virtually blind for its disadvantages and ignore/dismiss competing languages (of which there are many). Based on this it is safe to assume that most of them will also dismiss D, no matter how good it is. And since I already argued that the language really doesn't add anything new, programmers of other languages will also not be inclined to swap language.
It is funny that all attempts to make a new language that looks like C++ but doesn't have its advantages end up looking like Java. Maybe the Sun guys got it right after all. It certainly is a nice language to program in.
Slashdot Mirror
I think I even registered a slackware install way back in '96. I had no idea the linux counter still was alive. In any case I think it safe to say that most people who installed linux over the past few years are not aware of the existence of the good old counter so it must be way of by any standards and removing 90000 unverified users won't do much good.
DSL works fine in the Netherlands. The problem in the US is badly managed telecom companies trying to revive their business using a silver bullet called DSL.
In the Netherlands the copper network is in good shape and the largest problem has been getting the local telecom switches converted (a process that is still not completed everywhere). In most of the larger cities people have a choice between cable and DSL. DSL tends to be bit more reliable but also more expensive and cable has a bad reputation mainly due to the fact that companies like @home are active on the isp site there. The competition between cable and DSL has stimulated quality improvements in both.
I've had my DSL connection for nearly a year now. Apart from some technical problems in the beginning, I've enjoyed a good connection and get exactly what I payed for. In any case, DSL and cable are of course a temporary solution until we all can have a fiber optic connection.
Of course in Europe, local telephone connections not for free (like in the US), so people are more likely to take DSL to save money. Basically if, like me, you want to be online a lot, DSL is much cheaper than a regular modem connection. In the US your local connection is for free so you can be online all day relatively cheaply.
You'd still be using the Java API that way so it wouldn't help you much. The whole point of migrating to .Net is that you end up using .Net. Transforming the syntax doesn't really help since both languages are more or less equal in capabilities.
MS knew exactly what they were doing when they broke compatibility (presumably they can afford the legal staff to read a license) and it had nothing to do with improving the product. They expected that they could get away with it but were proven wrong.
They did implement AWT BTW. The problem was the way they called native code (i.e. they didn't use JNI). The only reason they did not support JNI was because that would allow them to create platform specific APIs in addition to platform specific implementation. MS always had the option to implement the necessary APIs but they refused to do so. Consequently, MS is entrirely to blame for the death of J++.
J# is Java without the Java platform (vm + API), except for the obsolete, partial jdk1.1.4 API. It is generally agreed that what makes Java so powerfull is the vm + API. The language itself is not very exciting really (your average OO language).
Decompilers are close to useless for maintenance activities unless you don't have the sourcecode (so either you threw it away, duh, or you're not supposed to have access to it). You do maintenance on sourcecode, preferably properly commented source code, preferably source code you are familiar with.
.Net or to bytecode is irrelevant. There's really only one certainty: migrating to .Net will generate additional maintenance activities since you will have to test whether it works and almost certainly you will encounter small problems. Once you have migrated and supposedly actually use the binaries you will find that you need to fix little things and even add new features from time to time. So you will keep tinkering with your old J++ code.
.Net languages. Also J# apps still use the rather limited 1.1.4 API and not the .Net API so it is really a tool to communicate with legacy J++ applications rather than to migrate those legacy applications (which eventually you may need to do after all). And once more: the whole problem of J++ applications being legacy software is because MS decided to abandon J++. First they charge you for a J++ development environment and now they charge you again for some crappy solution to let you keep using your J++ stuff. Probably the reason you decided to use J++ in the first place was to be compatible with the other wonderful windows stuff you paid for. You were screwed three times! If I had used J++ in the past, I'd be very pissed off. This also shows that you have to be careful to invest too much into C# or Visual Basic or .Net because MS may suddenly decide that they need some extra revenue and change things so much that what you invested in heavily suddenly becomes legacy software. Visual Basic changed substantially during its existence and generally 'migrating' apps between different versions cannot be done fully automatically (i.e. it will cost you). Just like with office it is highly debatable if the changes are worthwhile and that breaking compatibility is really necessary but you have no alternative so you upgrade and revise your old stuff.
Whether you compile to
It is worthwhile to mention that J# is not really a migration tool since there is no conversion of source code taking place into one of the typical
jdk1.0.2 code still compiles on jdk1.4 (ok, minus some minor API changes which were necessary and are generally easy to fix). The resulting bytecode can still be executed on a jdk1.0.2 VM. Try to do that with VB applications written in 1995 (when jdk1.0.2 was released) on the latest version of VB and you will see my point.
I keep reading about this on sites like this. It should be pointed out that even the JRE (java run-time environment) for jdk 1.4 is well below 10 megabytes (mainly depending on the platform). Of course if you download the full jdk you have a bigger download, mainly because that includes, among others, various tools and the source code for most of the API. But even then we are talking about 30 or 40 MB.
Go check out Opera or netscape, both have an optional download for the JRE 1.3.x. I think it was about 6MB. The JRE includes everything you need to run Java applications. It is hardly bigger than MS jvm and does a lot more.
Incidently, there is currently a beta of an enhanced 1.3.1 JDK that includes an activex component that fully replaces microsoft's JVM. Yes that's right, you can now run all your applets in IE using jre 1.3.1. Of course it doesn't support the MS specific extensions of the JVM.
Even as a migration tool its use is limited to projects using the partial jdk1.1.4 API MS supports. Basically any serious Java project nowadays is written using the Java 2 (i.e. 1.2 and upwards) API. And if they are written to the 1.1 API they likely use jdk1.1.8 rather than 1.1.4.
.Net objects) that don't require recompilation.
So basically it is a migration tool for J++ applications. Considering that is a MS product, it makes you wonder if this is the best MS can pull of after all the sole reason migration is needed is because MS decided to drop Java support (so they already screwed you once).
And even for J++ it is limited since it only allows you to compile source code, you lose information stored in e.g. forms, project files and so on. In addition, Java objects tend to be closesely tied to the Java API and reusing them basically means you are using the Java. So you might as well go for the real thing (including richer API, better performance and so on) and use the com bridge to communicate with the objects.
In any case, if this is just a migration tool, MS is going through a hell of a lot of trouble to present it as a Java alternative.
Some advise to people considering to use it:
- After 'migration' it is still Java code you are using. It won't be much faster and you will still have to maintain it.
- MS is not very Java friendly, they might drop support for the migration tool at their convenience. The long term strategy is C# (if it ever takes of), not J#.
- Migration to real Java is probably much easier unless you heavily rely on MS specific APIs
- There are ways of letting Java objects talk to COM objects (and consequently also
The whole problem is a simple economic problem. In the old days communicating peer reviewed papers to colleagues involved getting a printed copy to them. It was not more than logical to use the services of publishers to take care of the logistics and naturally they were in it for the money. In addition the publisher needed copyright in order to be allowed to print.
:-).
Nowadays there is an alternative means of getting the scientific content distributed: the internet. Many scientists have already discovered the internet as a means of distributing and obtaining papers. I can't remember the last time I actually had to go to a library to look up a paper. The internet totally eliminates the need for a publisher and could potentially greatly reduce the cost involved with publishing articles without affecting the level of quality of peer review in any way. In addition the internet provides additional possibilities in the form of for example moderated forums, mailinglists etc.
Unfortunately the publishers have built up an enormous power and are very reluctant to release it. Scientific journals are very profitable since they require a minimum of investment on their site and they have a small but loyal audience who are willing to pay relatively large amounts for journals. However over the past few years there has been a pressure to move to an internet based distribution of papers and journals. Despite this publishers still receive enormous amounts of subscription fees. At our university department, the library sucks up obscene amounts of our budget for online subscriptions and regular subscriptions. All this money (mostly coming from our government) flows to the publishers who add little or no value to the scientific product since they completely rely on external input (voluntary peer review, editing and academic authorship).
Most journals of significance in my field are either associated with ACM or IEEE, both of which are non profit organizations, both of which are charging significant amounts of money for membership and access to papers. A step forward would be if these two organization would abandon the publishing process and focus on internet publishing only. If those two organizations set the necessary steps, the rest would have to follow. As an author who has contributed to several IEEE conferences and who has even been a IEEE member (I have resigned because of the outrageous lack of value added by a membership) I would very much like to see this happen. After all they are supposed to represent our interests rather than work against us.
Basically I would like to see two things:
- Authors should retain copyright over their articles. The arrangement proposed for the ML journal seems reasonable, perhaps some GNU like legal document could be created to formalize it.
- Papers should be available and searchable at no cost. Currently I have to pay to be able to be able to download my own articles from the IEEE. I find that offensive (luckily I have copies on my homepage
In addition I would also like to see some innovations:
- XML based structure for papers to enhance searching.
- Addition of the before mentioned forums.
- distributed storage of content ala freenet.
If these requirements are met (even minus the innovations), I'll gladly renew my IEEE subscription and sign up for the ACM as well. I'm fully confident that with the logistics of the physical publishing process eliminated the membership fees can be reduced significantly. Alternatively this money can be used to set up more conferences and finance promising researchers.
However, currently I feel that neither of these organizations are working in my interests.
By definition, infrastructure should encapsulate every feature that is common to the applications that use it. The problem with UNIX is that such infrastructure is lacking. This has lead to a situation where in order to be able to compete with ms windows (which does have such infrastructure), applications have to include everything and the kitchen sink. Look at star office, it comes with its own widgets, its own printing subsystem, its own way of embedding components and until the recent beta it even came with its own desktop. Sure it is integrated with Gnome/KDE to some extent but since it also needs to be portable across the two it contains a lot of duplicated functionality.
KDE developers have understood this and are currently working to deliver such an infrastructure. Ignorant critics complain about konquerors ability to be both a browser and a file manager. However, once you understand what infrastructure is supposed to be you recognize that same ability as a good working infrastructure. File managers and browsers have a lot in common and therefore you might as well integrate them so that you don't have to invent the light twice.
The UNIX philosophy is to make something small and only once. Most unix applications only meet the first part of that philosophy and have to duplicate everything and the kitchensink because it is either not present in the infrastructure or not consistent enough to allow for easy integration (this should also be facilitated by the infrastructure).
I simply use visio to do drawings in framemaker. Works fairly well. I actually like framemaker a lot. It's a complex application with a lot of features which at the very least are non-trivial. However, once you get into it it seems like the people who built it anticipated what you really need.
Object linking and embedding supports both types of links you are talking about (even though it defaults to embed rather than link). On windows star office just uses the windows infrastructure so it is not different from ms office in that respect.
I have been reading very positive reviews about SO during the past few days. So, I decided to give it a try myself. Here's an overview of my experiences.
:-).
Let me start by saying that it looks promising, despite some obvious flaws which I will detail below.
Installation
I installed it on windows 2000. The installer looks pretty and userfriendly and seems to be doing what it is supposed to be doing in a straightforward way. I haven't tried the uninstall yet (duh) but I trust that will be equally good.
Compatibility with office
I normally use framemaker to write my papers but occasionally I have to deliver word documents. Star office certainly supports word better than framemaker. However, I found that it had some trouble with the word documents I had. All of these documents are rather complex and they use tables, crossreferences, images and paragraph styles. I was impressed that the document was successfully imported. However, there were a few errors that though easy to correct would have lead to visible errors if I had just printed them without looking at them. I had a similar experience with my powerpoint documents. Near perfect but not entirely perfect. Conclusion: you will need to revise imported non trivial office documents to make sure everything is the way as intended.
User interface
From the point of view of features most of it seems to be there and functioning. However, it is far from pretty. One of the key features of office is that it offers a polished userinterface. Obviously MS has a few graphical designers who know what they are doing. I can't say the same about SO it all looks rather primitive when compared to office (far better than framemaker though
In addition to the look and feel I found some of the dialogs a bit confusing. Some buttons don't have tooltips and there are slight inconsistencies, especially in the more advanced dialogs.
Features
After having played with star writer a bit I would say that I would actually prefer it over word and may even consider to use it as a replacement for framemaker in the future. As mentioned before, I use stuff like crossreferences and paragraph styles quite heavily. The prime reason I am using framemaker rather than word is because word is designed for stupid users and tends to "improve" the deisgn of your document on the fly which is a PITA if you do know what you are doing. Aside from that it's handling of graphics is really bad and you are at a constant risk of seriously messing up the layout of your document. Star office doesn't seem to suffer from these flaws and in addition has a build in literature references database!!!! That last feature (once I figure out how to use it because the GUI is a mess) might pull me over. Also I like the fact that everything is stored as XML and I am looking forward to any add on utilities that rely on this.
So in conclusion, star office looks very promising. I have only looked at star writer and compared it to word 2000. Word 2000 arguably is more polished and user friendly. In terms of features the two suits can compete even though star writer is lacking some of the popular word features (e.g. grammar checking). However, it also has features that either work better than or are not even present in word (e.g. the bibliography tool).
It looks like I am going to give it a try but it also looks like I won't be recommending as an alternative to office to less advanced users. If you know what you are doing and are equally annoyed with word as I am you might actually like it (otherwise I recommend you take a look at framemaker). It is not the office killer it is advertised to be but it certainly looks like you could use it for most of the stuff you would MS office for, especially when handling more complex documents. It's biggest problem is the user interface which just doesn't have the same quality as MS office.
Encryption is rather different than a gun in a few respects:
1) - the tools are software: duplication is easy. Guns are hardware and sophisticated knowledge is needed to make them.
2) - the algorithms are well known: you can make your own tools (without the backdoors). Building your own guns is a bit harder (though not impossible)
3) - there are open source tools (you don't even have to go through step 2 to obtain tools free from backdoors). Although the US occasionally hands out guns (e.g. stinger missiles to the afghan resistance a.k.a. taliban in the eighties), in general selling arms is profitable business.
Now about guns: you need a gun + an idiot to pull the trigger to kill people. Both prerequisites are available in large quantities in the US. In western europe, guns are a bit harder to get so we have less casualties as the result of guns (check the statistics if you like). Obviously, removing guns from society helps reduce the amount of people dying from guns. Doing so is a problem in the US however since billions of guns have been sold there in the recent centuries. So if you are in the US you are fucked, people around you are nuts and have easy access to guns. One day your nice neighbour or colleague may have a bad day and pull his guns on you (which he can buy legally and keep in his house).
Now lets turn to the real issue: why is the US pushing backdoors in encryption software: industrial espionage. Being able to tap in on information banks and businesses exchange throughout the world is very profitable business. A terrorist will just use illegal/free tools (probably on a illegal version of win XP or whatever). If there's one thing you can be sure of: terrorists don't like the US and they are not bloody likely to stimulate the US economy by actually paying for software produced in the US. What do you think? Bin Laden will actually log on to MSN and chat with his colleagues??? Come on!
The US government is using this situation to rearrange the world to make it a little bit more comfortable for the US leaders. Aguably the WTC tragedy was the best thing to happen to them in years. Some impopular anti-terrorist/anti-human rights laws can be pushed through. Suddenly they can be friends with Pakistan (a few weeks ago still referred to as a rogue state that we should be protected from by a missile shield). Everybody turns a blind eye while they whipe the Taliban of the earth and even Khatami is suddenly being friendly on behalf of Iran. In addition some former Soviet republics who happen to play an important role in producing and transporting oil are also the US' best friends.
It is touching to see all this friendship bloom. Unfortunately it is at the cost of millions of innocent Afghan civilians, already in big trouble because of the previous civil wars. What happened to New York was bad but the opportunistic way the US government is dealing with the situation is sickening.
What MS did is remove the ms vm and disable netscape style plugins. The old sun plugin was a netscape style pplugin.
Sun now has a beta version of jdk1.3.1 (in their early access program) that replaces the MS vm entirely. It is an activex plugin and it will even run applets defined in applet tags.
If you go to www.javasoft.com/jdc and sign in (for free) you can download it. You'll notice that sun has put up a call to testers to test it with existing applets in order to find compatibility problems.
It works great with IE 6 and I haven't found any applets yet that I can't run. It's stable too, no crashes so far.
"and just doesn't fit in with all the other UI apps I use (it won't follow themes or UI conventions). "
It's funny that you should mention this in a discussion on gui applications for linux. If there's such a thing as a standard look and feel for linux I have yet to encounter it. There's several desktop environments each of which come with their own widget sets, their own way of theming them, their own component models (if any at all) and their own look and feel. Generally you need all of them in order to run common desktop applications. There's no way you can target all those environments as a programmer. And applications written for one environment integrate extremely poor with the other environments (beyond the point of being able to display the user interface).
With Java you want to abstract away from it all so that it works on all platforms. That means you can't rely on native things to work consistently everywhere.
Limewire has achieved that. It's a simple, elegantly designed UI that works the same on each platform. Most of the native competitors pale in comparison and look clumsy when compared. It being crossplatform is vital since gnutella works better if there are more hosts that share files. The limewire people just have to design the GUI once and can focus on adding new features (which they do).
Admittedly there's a problem with integration with the native platform. However, on linux it is absolutely unclear what exactly this native platform is. Should sun integrate the JDK with Gnome, with KDE, with motif with X? Should they create separate jdk's for each environment? What about versions of each environment? The problem is that there is no standard and consequently all sun can do is target the most common denominator. They don't have that problem on mac os X or win32. The JDKs on these platforms generally integrate much nicer. They use file dialogs, the printing facilities, the native 3d, 2d and multimedia libraries, the clipboard and so on. Achieving the same on linux is nearly impossible since there are multiple implementations of each of those components. However, that is a linux problem and not a java problem. IMHO this is the primary reason that linux on the desktop is still not happening outside the developer community. Also I am very pessimistic about these issues being addressed in the near future.
You are of course right. The system is not democratic. Particularly the way of counting the votes and the procedure (or lack thereof) when a conflict arises is contributing to this. The only democratic outcome of the previous election would have been a reelection. Now the outcome of the election was determined by a puppet of the brother of the guy who won.
Yes it is well on its way to become a policestate. Now under the rule of an arguably undemocratically elected president (at least I would have quite a bit of trouble explaining the outcome to the ancient greeks who invented democracy), America is taking one step after another against its own people, who incidentally are cheering (talk about misguided people).
If all this would be limited to just the US I couldn't care less but unfortunately it isn't and europe is pretty much following the US in everything it does.
The recent events, however tragic, make it painfully clear what is wrong with the US. Within minutes after the crash the media machinery puts the video images in an infinite loop feeding them to their public, of course commercial breaks are inserted at regular intervals and CNN is likely to make significant profits in the next few weeks. Soon after, the US president, after being lost for a few hours in some distant US state, appears to make what can only be explained as a christian fundamentalist statement by calling for a prayer. The worst part is that this statement is fueled by opportunism (gotto keep the christian right wing people happy otherwise gore takes over in a few years) rather than true belief.
Now further opportunism dictates to nuke those damn arabs. Never mind the millions of refugees, never mind that the amount of innocent people who will die as a direct consequence of the media show the US army is about to give in the middle east will vastly outnumber the poor souls who lost their lives in new york (btw. a substantial amount of those people were muslim). And never mind that the short attention span of the US media and its audience is never going to outlive the chain of events about to be triggered in the middle east.
The days that reporters reported wars like in Vietnam are long gone. CNN is already being banned from Afganistan and soon the US military can (and will) do whatever pleases them. The gulf war wasn't as clean as CNN wanted us to believe and nor will this war (a b52 is not what I would call an anti terrorist weapon).
Incidently, I recall that a few months ago during the election campaign dubya had some trouble answering the question who the president of pakistan was. I hope his knowledge of this area has improved somewhat by now.
For the time being, enjoy the show just be aware that reality isn't what they show you on TV.
it's not the same thing, trust me on this. The google toolbar is very cool:
type a keyword and
- search the web
- search the site you are looking at
- search the google web directory
- search the archived news groups
- highlight occurences of the terms in the page you are looking at
In addition:
- a google button is present with configuration and links to google, advanced search and so on.
- a ranking is given for the page you are looking at
- a convenient up button is present that moves you one directory up (it's actually a dropdownmenu too so you can select any directory in the path)
- there's a button which shows you the directory in google web directory the site you are watching is indexed in (very handy for finding related material)
- and finally there's an information button which hides useful features as automatic translation, a link to google's cached copy of the page, similar pages and links to pages linking to the page you are watching
To the best of my knowledge opera only offers the first feature (searching the web). After bookmarks, the google toolbar is the single best productivity feature in my browser. It's a real time saver and it unleashes features that you'd otherwise never use because it's too much work. Often google's site search produces better results than the local site's own search option (usually some dumb altavista like engine).
Of course you are right. I was merely claiming that 0 bugs means nothing and is more a statement of testing incompetence than a statement of programming quality. Software with 0 bugs does not exist and good programmers are aware of that.
Of course good design, good methodology and so on helps improve product quality. Bugs found per kloc is certainly a good measure. However you should consider that something like mozilla is measured in MLOC (million lines of code). By your own measures there would be about 2000 bugs in that code. Given that the complexity rises exponentially rather than linearly if the size of a project increases that is probably a very optimistic estimate. I.e. even if all programmers are excellent (very unlikely), mozilla should be full of bugs waiting to be found.
To the credit of Mozilla they manage to track all bugs that are found in a very large development team.
Mozilla's attitude is much more mature than the attitude of e.g. Microsoft (deny until exposed in public) and ultimately leads to better products.
Unfortunately there are very few of these god-like programmers around. Reality dictates that you need a safetynet in the form of a good testing procedure to ensure product quality.
Bugfree software does not exist so the more bugs you find the better your tests are and since tests are also written & performed by programmers this is also a good indication of programmer quality.
A test that would result in 0 bugs found is worthless because that just means that the persons performing the tests are clueless. You don't test to prove that your program has no bugs instead you test to find the most visible bugs. In this respect, testing is very similar to research since you generally should try to disprove a hypothesis (and by failing to do so conclude that it must be valid).
Well, my complaint isn't that there aren't innovative projects (because there are) but that many self proclaimed OSS advocates are mainly concerned with reinventing the wheel and actually have a very conservative attitude towards anything new.
Just to provide some opposition:
ReiserFS. Check out Oracle's and MS plans to replace filesystems with databases. That is innovation, what Reiser does is just an admirable effort to duplicate the journaling feature invented by others.
Berlin. This is a nice project, however, apple has a vector based (pdf) backend for their GUI in mac os X. That is what I call innovative.
Perl, Python, Ruby, TCL, Java (the language, not the VM) are all variants of the same language constructs. Lots of syntactic sugar, good libraries. A lot of hard work went in them. But take a look at intentional programming from microsoft research or multidimensional separation of concerns tools from IBM and you will see my point because that is true innovation.
Some of the stuff I mentioned is actually open source BTW but I'm not discussing the license but the process and the community instead.
Both these projects I would refer to as rather marginal efforts (interesting though). Squeak is essentially a smalltalk clone. I haven't heard of jazz yet (will check it out soon).
I have heard of Alan Kaye. In fact I was about to listen to a keynote by him only last week(unfortunately it was cancelled at the last moment).
It's just C compiled to bytecode. It doesn't have most of the features that made Java popular (e.g. security, dynamicity, reflectiveness because inherent design decisions in the way C programs work inhibit this). In other words it is no replacement.
...). It is very rare to see an open source project that does not just duplicate features but instead introduces radically new features and paradigms. There are some research projects that use open source to distribute their stuff but these generally play only a marginal role in the open source community. The big open source projects are all about duplicating and imitating the bigger/better (in most cases) propietary counterparts.
Don't understand me wrong, maybe integrating such a thing with e.g. the gnu compiler would be useful. It would be great to create a single set of binaries and distribute them to the hubndreds of unix versions and other operating systems. But it has to be clear that this thing is orthogonal to java rather than the same.
In any case, I'm not impressed and even bored. Java is a relatively simple language. Most of the language concepts are easy to grasp, even for novice programmers. Yet people keep comparing it to more primitive languages such as C and suggesting alternatives which are basically C++ improved rather than Java improved. After years of being confronted with a rapidly growing java community, even some C++ programmers begin to appreciate things like garbage collection (after having dismissed it for years based on performance concerns). However, beyond garbage collection they are still largely missing the point (i.e. C is not so cool after all).
What I would like to see from the open source community (perhaps as a proof of concept) is a more ambitious effort, not just a (partial) duplication of features inspired by ignorance rather than innovation. Java has room for lots of improvement, lets set it as a baseline rather than a design goal. Duplicating all of Java's useful features should be a minimum requirement and not the ultimate design goal.
This post is rather harsh, I know. However, I think this is a fundamental problem with open source in general and linux in particular: it's mass production of commodity software components (kernels, compilers, IDEs, word processors), not creation of new ones. If cutting edge technology is your thing, the propietary world is still the source of it (speech recognition, cool 3d technology, AI related improvements to user interfaces, compiler technology, languages,
In short I find the open source community boring & backwards. Its contributions are very useful and I use them on a daily basis. However, as a researcher I don't want to wait for thirty years to see things like OO being grudgingly accepted; I hate it when linux is advocated as a windows alternative when the people doing so largely fail to even realize why windows got popular in the first place. The same applies to Java and Visual Basic (yes, from the *evil empire* you understand me correctly). The features that made those things popular have yet to be duplicated. Effords like the one advocated in this article are just so backwards they only make me angry. Go do something useful! Don't waste your time figuring out how to reinvent the wheel, I already have half a dozen.
How about a killer app? The problem is that right now if you start using ipv6 you are pretty much alone. Actually you might as well unplug your network cable, since you won't be able to do much useful stuff with it.
What is needed is ipv6 only services (e.g. mp3 peer2peer filesharing) AND an easy way to get an ipv6 number for your clients/servers that can coexist with your current ipv4 number (i.e. your computer has both an ipv4 and ipv6 number). The easy part is essential because that prevents that people start creating ipv4 gateways to such services (thus removing the need for getting an ipv6 number). There are plenty of ipv6 numbers available so getting and registering one should be made as easy as possible (something like a distributed, global dhcp server that would automatically get you one based on your mac address would come in handy). Come to think of it, why not just automatically convert those mac addresses into ipv6 numbers (mac addresses are supposed to be unique anyway but I'm not entirely sure this is a great idea)
As I understand it, ipv6 can be tunneled over existing ipv4 networks, so it shouldn't be a problem if some routers inbetween ipv6 hosts are ipv4 only.
This would cause the amount of client pc's with ipv6 numbers to gradually grow. Also since lots of PCs don't have static ipv4 numbers, the amount of servers on ipv6 would also grow. Eventually, there will be a critical mass of ipv6 servers and clients and the switch can be made.
Currently there are a lot of p2p applications in development. I imagine, implementing such stuff would be a lot easier using ipv6 with its improved features. Another killerapp could be streaming multimedia (you want to see this great movie, get yourself an ipv6 number now!!).
That has never happened to me and I visit the CNN site on a daily basis. I distrust CNN for another reason: they are a little to friendly with the US government and often act as a PR machine for the US military.
CPU's and memory are cheap, good programmers are expensive. Your assumption that speed is essential is correct but it applies to time to market more than to performance. Execution speed is not so relevant since hardware is relatively cheap compared to programmers (even lousy ones). The year salary of one programmer will likely solve any performance problems you have when converted to extra hardware.
Swing is not lightweight but given a decent PC it will run. In fact all development tools I use are written in Java and I'm not complaining about performance. I'm sorry I can't show you a java program that is faster than the C++ version. However most java applications I could show today, you would still be in development if they were written in C++. That's why java is picking up, you get things done in it.
That's also why Java became popular on the server first. Typically server software is "one of a kind" meaning that each month spent on developing the app adds to the cost. If you have a small development team of say 5 persons costing about 5K $ (good luck finding them that cheap) and you can save 1 month that translates into a lot of hardware. Java development is generally seen as somewhat faster than C++ development so that translates into huge savings on one of a kind software development. Also Java is very scalable, it literally can run on anything from a credit card to a mainframe, so if you have a performance problem, throwing more hardware at it will likely make it go away. And since you develop faster you have something extra to spend. That's why many people creating server software have adopted Java. It works, it works reasonably fast and it scales nicely.
I read the specs. 20 years ago it would have been impressive, but now...
.Net for a very good reason: it adds a great deal of flexibility. Java has already shown that VMs can efficiently execute numerical code. HP has shown that interpreting bytecode can actually be faster than compiling it. There is no sound reason to dismiss the concept of a VM anymore. Java vms are even available on embedded machines now so size is not an issue either.
.Net, the JVM or some other vm (unix currently lacks one) so that you can at least take advantage of the features and libraries offered by such environments.
Lets summarize:
- no generics, yet (he was 'thinking' about it)
- I haven't spotted an interface construct like in Java, very useful and no performance penalty -> include it!!
- no dynamic classloading, another useful Java feature
- no reflective features, another useful Java feature.
- no new features worth mentioning
I would propose this language were called Java--. It removes features the author of the spec deems irrelevant and doesn't add any new ones. Java has its flaws but these flaws are tolerated because it also adds useful features.
Incidently, the VM approach has been adopted by MS now in
If the author of the spec still wants to go ahead with implementing D (aka Java--), I'd strongly suggest to make it compile to
And finally, there is no market for this language. C/C++ programmers are generally so much in love with the language that they are virtually blind for its disadvantages and ignore/dismiss competing languages (of which there are many). Based on this it is safe to assume that most of them will also dismiss D, no matter how good it is. And since I already argued that the language really doesn't add anything new, programmers of other languages will also not be inclined to swap language.
It is funny that all attempts to make a new language that looks like C++ but doesn't have its advantages end up looking like Java. Maybe the Sun guys got it right after all. It certainly is a nice language to program in.