From: Chris Davies <c.davies@cdavies.org>
To: gentoo-dev@gentoo.org
Subject: Re: [gentoo-dev] JRE support - is it worth it?
Date: Tue, 20 May 2003 17:32:45 +0100 [thread overview]
Message-ID: <20030520163245.GA6672@vaughan.foofalicious.com> (raw)
In-Reply-To: <200305192349.32237.absinthe@gentoo.org>
Hi,
I for one do not wish to see JRE support dropped from gentoo, and have a number of reasons why.
Having installed gentoo over 56K modems a number of times, I would have found it immensely frustrating had I had to install a JDK just to get Java support in mozilla and it's derivatives. We are talking 5+ hours for a JDK, compared to only 2 for a JRE. Now I have broadband, it is easy to overlook those people who can't get fast connectivity, but I don't think they should be neglected merely for the sake of simplifying a supporting script. The majority, I suspect, only have any Java product on their machine to get the java plugin.
Secondly, there is simply no advantage to compiling java applications on the system in question. It is a complete waste of time. Java compiles into machine independant bytecode. The same Java code compiled by the same compiler on two different architectures should produce exactly the same result. So unless you plan to offer gcj as an alternative compilation tool, compiling from source is utter waste of the user's time. It is dogmatic in the extreme to suggest that because we compile programs that generate machine dependant object code on the system itself, all programs on the system should be compiled. Compilation is a means to an end, that end being object code that is more efficient than offered by binary distributions. In cases where the build process is long or difficult, often binaries are offered (openoffice-bin and phoenix-bin spring to mind), so saying this is a source distribution is false. I firmly beleive Java packages should be offered as binary until gcj is fit to compile them, and that one or other of the JREs should be the default Java environment.
It is true that some packages do require a JDK, like Tomcat, but I can't see that as being a reason that all Java packages must require a JDK. It should be pointed out that Tomcat is distributed as a binary, so the JDK is only a runtime dependancy. Why inflict the JDK on users who neither want or need it?
Well, thats my rant of the day over with :)
Thanks,
C.Davies
* Dylan Carlson (absinthe@gentoo.org) wrote:
> Greetings,
>
> This is not an RFC, per se, but I am posting this with the general purpose
> of soliciting discussion. This might be a controversial subject, so I
> think it warrants some conversation before it reaches the level of an RFC.
>
> I have talked with a couple of people the idea of dropping JRE support in
> Gentoo. The reasons this idea came about:
>
> 1. Gentoo is a source-based distribution. We have Java-based packages
> that build from source for the same reason that have C packages that build
> from source. Yet, it's unthinkable to have a system without a C compiler.
> Thus it should be almost as unthinkable to have a system without a Java
> compiler.
>
> 2. Few people actually use a JRE inside of Gentoo for the above reason (if
> any) ... yet of course reliable statistics on this are not available.
> Furthermore, some packages require a JDK to run, namely servlet engines,
> development environments, and rare apps that depend on the JDK without
> actually compiling/debugging anything.
>
> 3. Jikes will compile most things, but not everything. Furthermore, Jikes
> lacks a javadoc tool, among other things. So it's not suitable to have a
> JRE + Jikes combination, though in theory that might be possible someday.
>
>
> User impact to the above idea:
>
> 1. Download & install size: Using 1.4 as our example, JRE's are about a
> 22MB download. JDKs are about 42MB. It's also safe to assume that the
> JDKs also occupy roughly twice the disk space, though I don't have
> specifics on it for this email.
>
> 2. Reduced complexity in java-config(1), simplifies the nature of Java
> ebuilds for both Gentoo developers and users submitting ebuilds alike.
>
> 3. Reduced QA burden.
>
> I would like to hear your thoughts on this.
>
> Cheers,
> Dylan Carlson
--
gentoo-dev@gentoo.org mailing list
next prev parent reply other threads:[~2003-05-20 16:31 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-05-20 3:49 [gentoo-dev] JRE support - is it worth it? Dylan Carlson
2003-05-20 16:32 ` Chris Davies [this message]
2003-05-20 18:22 ` Todd Berman
2003-05-20 22:41 ` Dylan Carlson
2003-05-20 23:47 ` William Kenworthy
2003-05-21 0:56 ` Chris Davies
2003-05-20 16:40 ` Riyad Kalla
2003-05-20 18:01 ` Brad Laue
2003-05-20 21:57 ` George Shapovalov
2003-05-21 9:48 ` [gentoo-dev] " Stephan Feder
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20030520163245.GA6672@vaughan.foofalicious.com \
--to=c.davies@cdavies.org \
--cc=gentoo-dev@gentoo.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox