[Savannah-hackers] Re: Invalid URL in registration response

[pcg]

On Tue, May 13, 2003 at 10:11:15PM +0200, Mathieu Roy <address@hidden> wrote:
> So defining a $variable is creating a "proprietary version"?

Not at all. But using illegal characters (lots of them) in URLs and
expecting them to work clearly is.

> I assume that you are using "a proprietary version of open standard"
> with your mailer. 

Your assumption is wrong of course.

Look, before you reply again and demonstrate that you have no clue, how
about reading the corresponding standards?

> Because you email address (<pcg(Marc)@goof(A.).(Lehmann )com>) is
> clearly not a standard address. My mailer, gnus, does not understand
> it and I have to copy/past your address from your signature to reply. 

Well, I use Internet Mail, which is defined by rfc822 (or rfc2822). If you
mailer has problems with it, I recommend getting one that does, in fact,
understand e-mail.

Of course my address is fully standards compliant.

Couldn't you just, and after repeatedly demonstrating that you have no
clue about the relevant reocmmendations or standards, read them?

> But let it end on this, address@hidden is not a place for
> flamewar. 

Well, I am and still just wanted to tell you that there is a bug
somewhere.

Obviously, you don't care about bugs. Instead, you invest a lot of time
and effort into educating me that you do not understand URLs and e-mail
addresses.

> You said a URL does not work. It does. 

A URL does work. The one in that mail doesn't.

> Type any of these urls, the worst you can get is the content after
> ?something= to be ignored. Which does not broke the system.

Actually, the worst I get is that my proxy or browser will complain about
a protocol error. And if I get a message form my proxy than this just
means that my browser didn't check the URL, which is bad, but ok.

Please, you do not know what a URL is, how it does look like etc. Please
finally go and read the rfc where it is defined. You will then hopefully
see that it does, in fact, not work, simply because it isn't a valid
url.

I really wonder how you get the energy of repeatedly claiming that the
URL is valid.

I mean, it is *extremely* easy to verify my claim. It takes much less
time than writing a reply.

Also, just fixing that bug will probably take much less time than writing
your ridiculous replies.

Fact is:

- there is a bug
- nobody cares for fixign it
- it's more fun to write emails with hilarious claims

That certainly isn't worth for that mail address. I recommend that
somebody else should get these e-mails, somebody who actually invests
his time into fixing bugs instead of re-iterating the same wrong claims
without reading the defining documents, or, heaven forbid, actually fix
the bug.

I mean, don't you think it's extremely anti-social to treat people who
just report bugs and want to honestly *improve* savannah.gnu.org in the
way you do now.

There is no problem if you don't have time to fix the bug, or need help,
or don't understand the issue. But obviously you do have the time to write
these e-mails.

Instead you are promoting a very negative image of gnu.org, namely "we
are not interested in fixing bugs, but we prolly do argue about them
till the reporter gives up".

Ok, I understood the message, I'll go somewhere else next time.

> This URL, Uniform Resource Locator, is something that was added in 2
> minutes to save time for  users that have to reregister a project

Well, it's not a URL. Do you have a problem understanding that? I guess
so... what's your point? Please go and verify in the rfc that it is, in
fact, not a URL.

> This is clearly not an essential thing. This is clearly not a
> "URL extension", it's just a non-cleanly formatted url provided

It's not a "non-cleanly formatted url". It's not a URL at all.

> just-in-case. If it does not work with a browser (it does not with most
> free software browsers, actually), you'll just have to refill forms.

Well fine. Sorry that I have reported a problem. It's of course
etxremely easy to fix, but you don't want that. Instead you try to tell
me that my e-mail address is wrong and that URLs have spaces in them,
and this is perfectly ok and not a proprietary extension.

Really, that's NOT the way to treat bug reports. It's not the way to treat
anybody. And it certainly is NOT the way that people would expect from
anybody from gnu.org.

> More complex implementation would require more time and since it's

It would probably take a few seconds to change the text. Even fixing the
URL would probably take a few seconds at most.

> absolutely not essential, it would surely not be done. 

I have to date not found a single browser that understands that URL.

> What do you want: to remove a feature that works for many persons?

Not at all. Why are you talking about that? I suggest reading the e-mails
I sent earlier, where I describe *exactly* what could be done.

I don't see a point in answering the same questions again and again if
you ignore my answers.

previous e-mails (your surely haven't).

> We wrote in this mail you talk about:
> 
>         "Some users find it useful to use the big re-registration URL
>         provided in the acknowledgment e-mail you received after
>         registration." 
> 
> It clearly shows that we know that this URL is pretty weird (the

Maybe it shows that to the author of that text, but to normal people
it just suggests that some users might find the following url useful.
After all, that is what is written there. I honestly fail to see where it
mentions that the following actually is not a URL, or doesn't work with
the majority of browsers.

> We have more complex issues to deal with, for instance sftp support,
> which is asked by about 50 users. So enhancing this URL is clearly not
> to me something important right now. 

Well, fine. But instead of telling me this and just fixing the text or
something else you invest lots of time telling me that I am wrong, my
e-mail-address is wrong and other - sorry to say it - idiotic stuff.

> That said, Savannah is a project. You can contribute if you want
> to.

I did contribute. I analysed a bug and told you about it.

> If you feel it's really something that need to be changed, send us a
> patch, we'll apply it. That's how free software work.

I'd probably do that (care to tell me where to download the script? I
can't find it, sorry, my fault).

I find it a shame, however, that gnu.org drives possible contributers
actively away by telling them their contribution is wrong and not wanted,
like you did.

I mean, no problem I'd go in and fix the problem for you. But since you
clearly told me that there is no bug and I am wrong, how I am I supposed
to know that, in fact, what you wrote is just some bad joke and you want a
patch?

> It's usually more appreciated when you're spend more than 10 hours on
> a free software project to get request that say "this does not work

I am definitely the wrong person to tell that - I certainly contributed
the better part of my life to free and gnu software. But regardless of
wether I contributed more, less, or whatever amounts of time on free
sofwtare projects - it still does not make sense to waste more of _YOUR_
time on writing these kind of drive-away mails than it would take to fix
the problem.

Your whole argument ends in: "it makes more sense to argue about an
obvious bug than to fix it". And wether or not I contribute to free
software at all does not make this argument true.

However, I do _not_ drive away possible contributors by telling them some
made-up ideas and claiming they are true.

If some other, less knowledgable, person came along and would get your
reply she'd probably just give up and never come back.

> for me, I would like it to work that way" instead of "this is not an
> URL, this is awful, this proprietary".

Well, sorry that I didn't get the tone right for you. However, I don't
think my tone was insulting or bad or anything. And I certainly didn't
react by repeatedly, and knowingly talking about subjects I didn't
understand at all.

I think you really owe me an apologize.

> The fact that a software fails to perform correctly on some conditions,
> or in some systems, or fails to comply current policy documents is a
> bug. It does not make of the software a proprietary one.

Of course. Did I ever claim otherwise?

_You_ claimed it's proprietary, since you repeatedly said that it is a
valid URL. Since it certainly isn't (I told you about that and rfc1738
right along), you are obviously referring to a propietary version of URLs.

In any way, you didn't serve free software well with your attitude and
this discussion. It's pure waste of time on both sides, and I am sure it
would have takne me less time to fix that bug than to argue that, in
fact, it should be fixed.

-- 
      -----==-                                             |
      ----==-- _                                           |
      ---==---(_)__  __ ____  __       Marc Lehmann      +--
      --==---/ / _ \/ // /\ \/ /       address@hidden      |e|
      -=====/_/_//_/\_,_/ /_/\_\       XX11-RIPE         --+
    The choice of a GNU generation                       |
                                                         |