10 comments
Comment actions Permalink

On Tue, 15 Mar 2005 07:42:53 +0000, Tom wrote:

...seems to be wrong:
http://cld.blog-city.com/read/1126337.htm


Note his @sun.com email address, add salt.

0
Comment actions Permalink

Er, that's not a myth. It was never fast, noone really claimed it was. (If you disregard marketingblurbage).

0
Comment actions Permalink

Note his @sun.com email address, add salt.


What he said. Swing on Linux has plenty of problems, see the previous posts about font antialiasing and input problems. That article is clearly publicity.

0
Comment actions Permalink

Clearly? The article names N faults with SWT and Eclipse, you're saying he's making them up? Swing having N problems and SWTEclipse having N problems, is not two mutually exclusive states. That article happens to be about SWTEclipse problems that aren't Swing-problems.

0
Comment actions Permalink

Right, and by not mentioning the problems that Swing does have, he's pushing the article into publicity IMO. I'm not saying that he's making facts up, I'm saying that the slant of the article is far from impartial, and it should be treated with caution. But I've not used Eclipse on Linux, so I can't comment with authority. I do know that SWT solves some of the Swing problems that really annoy me with IDEA on Linux (font rendering and native input).

0
Comment actions Permalink

In article <30936084.1110898738546.JavaMail.itn@is.intellij.net>,
Colin Fleming <no_mail@jetbrains.com> wrote:

Right, and by not mentioning the problems that Swing does have, he's
pushing the article into publicity IMO.


I must disagree - he is pointing out real problems, and I really do not
see where he has to point out Swing flaws to be fair - the eclipse folks
don't.

I use Eclipse and IDEA on MacOS X, and IDEA performs marginally better.
I suspect that this is partially due to JB efforts, and partially due to
Apple optimizing Swing over time.

The problem? SWT is used by fewer people on the Mac, and has a smaller
group contributing to it. As time passes, it will likely get better,
but it is not clear to me how fast it will get better.

Scott

0
Comment actions Permalink

So if you compare two things, you must always make sure not to list more negative points for one of the items?

I don't need you to protect me from articles you find biased. I can think for myself. I know biased articles, that article doesn't come across as one.

The thing with SWT is this; if it were to offer everything swing does, with the same quality/speed/footprint/ease of use and all that, then it's still virtually useless since Swing support is in all jdks. For SWT to have any place what-so-ever, it needs to be of clearly better quality/faster/smaller footprint/easier to use and all that. Why else would you pick it instead of Swing?

Swing's pros and cons are irrelevant in this context since SWT doesn't even match up to the cons. As the article said: give SWT another year or two, and it'll be just as good as AWT... seven years ago!

0
Comment actions Permalink

shrug

So use Swing. I'm not trying to change anyone's opinion here, just stating what I thought of the article.

0
Comment actions Permalink

I think you're missing my point. This is not about what I will and won't use. My original post was in response to you claiming the article was biased based on it ripping on SWT. And that's just silly!

There's no need for minute justice within one and the same article, the article in and of itself is negative critique directed at SWT, as such it plays a balancing role against all the marketing lies regarding SWT's (and to an even bigger extent: Eclipses') qualities. The points in the article are not made up.

0
Comment actions Permalink

And I think you're missing my point. I never said he'd made up anything about SWT or about Swing, I said that the fact that he works for Sun, does nothing but bash SWT and Eclipse and devotes the last 5 paragraphs to promoting ways to switch to NetBeans makes it essentially advertising. He doesn't, for example, suggest that people switch to other Swing-based IDEs like IntelliJ. He also provides no actual facts or measurements to substantiate his argument - quoting messageboards on the internet is the lamest piece of pseudo-justification I've ever seen. It's like an article posted by someone working at Microsoft comparing J2EE to .NET using Rolf Tollerud's posts on TSS as references. The only remotely valid piece of critical information in the article is the Eclipse bug report.

Looks like we'll probably have to agree to disagree. If you liked the article, great. I thought it was rubbish.

0

Please sign in to leave a comment.