|
Let Java Go.
I think it's a very well written article by Eric on why Sun should loosen it's control on Java. He explains this by giving solid examples.
Check http://www.catb.org/~esr/writings/let-java-go.html
It makes me rephrase Joel's wordings in one of his article, "It is very important for you to decide whether you are creating APPLICATION or creating PLATFORM."
JD
http://www.phpkid.org
JD
Sunday, February 15, 2004
Well said. I second the call!
:-) stw
NotesSensei
Sunday, February 15, 2004
I believe they will annouce the open sourcing of Java just after the press conference by MS on the open sourcing of Windows.
Craig
Sunday, February 15, 2004
one thing i like about Java being owned by one company is that you do not have forks and different variations. i'm not talking about J2SE, J2EE, J2ME. i mean, there aren't 4 different versions of the API for the entriprise level stuff that is currently handled by J2EE. i know there are container specific stuff, but overall Java is Java.
michael sica (michaelsica.com)
Sunday, February 15, 2004
The ingratitude of these open source advocates is just unbelievable. Sun has done more for open source than anyone. You probably have to go back to the Medici to find a company that has freely dispensed as much intellectual capital as Sun. In every way that counts Java is already open.
MT
Sunday, February 15, 2004
Eric destroys any credibility he might have had by comparing share price instead of market capitilization, thus showing he has no idea what he is talking about when he implies Red Hat is doing better than Sun in the market.
The rest of the article is hardly better: dragging out a SUN business move from 1984 (the NFS standardization) and implying, without backing it up, that it is analogous to releasing Java in today's business environment is just weak.
Sorry, this is not a particularly well-written or compelling piece.
Mike Treit
Sunday, February 15, 2004
I think that open-sourcing would weaken Java (and invite Microsoft to 'pollute' it, as were trying to do). It is really important that there is a standard to which implementations must conform in order to be considered Java.
The open-source model is great in certain situations, but what it has so far failed to deliver (in general) is good user interfaces and good documentation. The good UIs and docs that exist for OS projects invariably come from corporate involvement. Unless you have the resources to do user testing (even most companies don't), or technical writers who can easily interact with the developers (technically possible for all via the net, but does it happen?), you're unlikely to produce a decent UI or a decent set of docs. How many open-source apps have you used where the Help system turned up a blank page, where there wasn't a manual available, and you had to rely on the goodwill of the developers? (Actually, ask yourself that same question about low-cost Windows software.)
Java's UI system (Swing) isn't great, but it isn't terrible either. You can turn out decent apps with Java. There is a wealth of documentation for it from Sun (just as there is for the MS languages on MSDN). I just don't think the open-source world would be able to deliver the goods in this case. You need a benevolent dictator.
C Rose
Sunday, February 15, 2004
The divergence of vendor-specific extensions and variations in J2EE servers is evidence that Java itself would get hijacked in different directions if Sun were to "set it free".
T. Norman
Sunday, February 15, 2004
Java needs to go the way of Unix. You can't call an operating system Unix(tm) unless it follows the Single Unix Specification. It does matter what vendor or even what original source it comes from.
There needs to be a Java consortium that handles the Java standard process. The problem is that currently only Sun technologies are "blessed" and real solutions (like IBM's SWT) are on the fringe.
The development of Java is already very similar to that of Linux; many companies and organizations working on Java technologies. But unlike Linux, there is no central neutral party handling the core.
Java cannot be set free like Netscape was set free as Mozilla. But it still needs to set free, in some form, to grow as an independent platform.
Almost Anonymous
Monday, February 16, 2004
"one thing i like about Java being owned by one company is that you do not have forks and different variations. i'm not talking about J2SE, J2EE, J2ME. i mean, there aren't 4 different versions of the API for the entriprise level stuff that is currently handled by J2EE. i know there are container specific stuff, but overall Java is Java."
BWAHAHAHAHAHAH!
MS JVM vs. IBM JVM vs. Sun JVM vs. Blackdown JDK all with subtle differences.
WebSphere vs. Weblogic vs. Tomcat vs. JRun all with subtle different extensions
On the other hand you have Perl and Python. Both are open source projects and have no problem whatsoever with forks. (There are a few forks for experimental purposes, but no forks trying to take over the leadership).
I highly doubt Sun is ready to open up Java, but I just couldn't help but respond to this misconception.
Richard P
Monday, February 16, 2004
"The ingratitude of these open source advocates is just unbelievable. Sun has done more for open source than anyone."
Don't be an idiot.
There are big problems with java right now, for example not being able to put it in a cd. Take the *BSD projects, it's probably the SINGLE ONLY package that you have to manually download.
It's quite silly. Sun claim to be free but aren't. The java issue for example. Or the fact that they refuse to tell people how their hardware works, which leaves projects like openbsd unable to write code to use features in their hardware. It's about action, not bullshit that you finding from CEO types, and on the ground, they're very non-free.
fw
Monday, February 16, 2004
"I think that open-sourcing would weaken Java (and invite Microsoft to 'pollute' it, as were trying to do)."
What exactly do you mean by 'pollute' Java?
And if Java was Open Sourced, how could MS handle (ie 'pollute') Java without handling Open Source software, thus admitting that all the bad stuff they said about Open Source was wrong?
And what kind of signal would it be if MS was emitting MS Java, while saying 'don't mind us going Java, just keep on going .Net and C#'?
And why hasn't MS 'polluted' Linux, which is already Open Source, if 'polluting' is such a great activity for MS?
Martin A. Boegelund
Monday, February 16, 2004
>"On the other hand you have Perl and Python. Both are open source projects and have no problem whatsoever with forks."
But are they as reliable as Java when it comes to cross-platform compatibility?
T. Norman
Monday, February 16, 2004
"Open source is hardly a zero-revenue model; ask Red Hat, which had a share price over triple Sun's when I just checked. "
On my planet things seem a little different:
Redhat: $ 3,209,010,600
Sun: $ 18,391,881,320
(closing prices 02-13-2004)
And both companies seem to be loosing money, not gaining.
Just me (Sir to you)
Monday, February 16, 2004
I could establish a 10 billion dollar company tomorrow, if I wanted.
How? Selling dollar bills for 50 cents.
Revenue is important, but not everything.
biznass whiz
Monday, February 16, 2004
You have 20 billion one dollar bills?
Cool...
Mike Treit
Monday, February 16, 2004
>>It makes me rephrase Joel's wordings in one of his article, "It is very important for you to decide whether you are creating APPLICATION or creating PLATFORM."
<<
Without getting into whether it is better for Java to be open source or not, what does it have to do with whether Java is an "application" or "platform"?
Something does not have to be open source in order to enable it to be a "platform"? In fact, if I recall one of Joel's examples, he cited the case that the first version of Windows had a redistributable runtime. (correct me if I'm wrong.) Now, that runtime was not open source at all.
A.F.
Avrom Finkelstein
Monday, February 16, 2004
"How many open-source apps have you used where the Help system turned up a blank page, where there wasn't a manual available, and you had to rely on the goodwill of the developers?"
While I agree in general I have to point out that Python is an exception to this rule. The documentation is amazingly complete and well-written.
Chris Nahr
Monday, February 16, 2004
I'd also like to point out that Linux MAN pages are very complete pieces of documentation. And not just for applications. You can "man someconfigfile" and get the syntax and other information in detail.
The most fringe the open source project, the more sparse the documentation. This is true for closed source projects as well. I think there just happens to be more fringe open source projects...
Almost Anonymous
Monday, February 16, 2004
"WebSphere vs. Weblogic vs. Tomcat vs. JRun all with subtle different extensions"
I don't think Tomcat is a J2EE container, it's just a servlet and JSP container. So it doesn't belong in that list.
And you said, different extensions. That's container specific stuff. Outside of the J2EE spec, no? If you want to be portable between containers, don't use container specific stuff.
I'm not a J2EE developers, so I could be off the mark. I'm still getting my feet wet with Java, and work as a Project Coordinator (after a couple of years as a Coldfusion Web Developer).
michael sica (michaelsica.com)
Monday, February 16, 2004
Justme:
Mr. Raymond was talking about share prices. Which makes him a complete geek.
Alex.ro
Tuesday, February 17, 2004
Alex,
you mean the price of an individual share? How is that related to the value (let alone the revenue) of a company?
Is this typical of OSS people that feel the need to advice companies on business models? Seems even worse than PHBs selecting development platforms based on the looks of the IDE icon.
Just me (Sir to you)
Tuesday, February 17, 2004
Funny as it is, Eric is writing about the price of one share.
I could start a company with 10 shares, $1000 a piece. Wow.
Alex.ro
Tuesday, February 17, 2004
Justme and Alex:
How much would you pay for 1 share in a zero-revenue company?
Martin A. Boegelund
Tuesday, February 17, 2004
If I could sell it at a profit? =)
Alex.ro
Tuesday, February 17, 2004
To whom? Justme?
Martin A. Boegelund
Tuesday, February 17, 2004
Martin,
that is a nonsensical question.
Just me (Sir to you)
Tuesday, February 17, 2004
Let me elaborate:
The price I would be willing to pay would be based on my estimates of the future share price as a delta from the current (the opportunity), my confidence in that prediction (the risk), the money available to me for speculation in that risk bracket and the martket alternatives.
e.g. Getting in on an IPO of a company before its first product launch at a price of X where my estimate of the price in 6 months would be 3X to 10X with a very high certainty would be most welcome. OTOH an opportunity of buying shares of a company with billions of dollars of revenu at price X, where my estimate is that the shareprice would vary between 1.0X and 0.8X over the next 5 years is very unappealing.
The current revenue is not a factor. Estimated future revenue (or better, estimates of future revenue differences from general expectation discounted for risk) can be.
Just me (Sir to you)
Tuesday, February 17, 2004
Justme:
You wrote that
"you mean the price of an individual share? How is that related to the value (let alone the revenue) of a company?"
indicating that the price of an individual share is unrelated to the value and revenue of a company. I was just curious how you would calculate a share price without taking company value or revenue into account.
You admit you might take future revenue into account. Good. So we agree that (expected future) revenue does play a role in the calculation of share price?
How about value of a company? Doesn't that part play a tiniwinititsytiny role in the calculation of the share price?
Martin A. Boegelund
Tuesday, February 17, 2004
Martin,
it is not that simple. It is the difference between your own calculations of the changes and as they might influence perception, and those of the rest of the market.
Current share prices already include the future. You as an investor speculate on a future change of perception of the future perception. In all that speculation there might be some links to changes in value based on revenue outcomes, but the links are far from simple (hey, if we figured them out we'd be rich, right).
Still, this has nothing to do with the original observation that there is absolutely 0 information regarding revenue or profit in the observation that share X costs x and share Y costs y and x>y.
Just me (Sir to you)
Wednesday, February 18, 2004
I am glad that SUN has control of JAVA. Please don't take my SUNshine away.
Nice Guy
Saturday, March 27, 2004
 Recent Topics
Fog Creek Home
| 
