LWN.net Logo

Hmm

Hmm

Posted Feb 26, 2004 22:04 UTC (Thu) by Ross (subscriber, #4065)
Parent article: The Luxury of Ignorance: An Open-Source Horror Story (catb.org)

I completely understand his frustration but he's a bit abusive. I'd also
like to point out that the broadcasting feature being turned off and the
listening address being set to localhost by default may not be the fault
of the CUPS developers but the "fault" of the distribution.

I'd also like to say that sending out periodic broadcasts is highly
annoying on a large network and is one of the reasons I don't like
Windows. Couldn't CUPS use multicast or maybe listen for a broadcast
request and only respond if they haven't received one in the last N
seconds? That way there is no traffic for uninterested computers or no
traffic when nobody is doing autoconfiguration.


(Log in to post comments)

Not so abusive

Posted Feb 26, 2004 22:20 UTC (Thu) by dang (subscriber, #310) [Link]

I don't think that he is being abusive at all. The piece reflects the very real frustration that users feel when confronted with UI and doco that are broken.

And his point is well taken. For most users, usability is real measure of success.

Really?

Posted Feb 26, 2004 23:07 UTC (Thu) by Ross (subscriber, #4065) [Link]

He calls the CUPS developers idiots and has nothing nice to say about
them (other than he likes the design of the network-based autodiscovery).

My point was that he's placing blame on them when the software might have
been reconfigured by the distributor without changing the documentation.

Really?

Posted Feb 27, 2004 21:49 UTC (Fri) by ren_123 (guest, #19852) [Link]

>He calls the CUPS developers idiots and has nothing nice to say about
them

Oooooh, the typical luserish behavior of 'I only listen to your
complaints if you give me a pet on the back'. That's almost comparable to
chantage.

Do you also start with positive, constructive comments if somebody knocks
your teeth out?



Really?

Posted Feb 28, 2004 0:41 UTC (Sat) by piman (subscriber, #8957) [Link]

Eric would have grounds to criticize the CUPS developers' technical abilities, had he implemented (or substantially designed) a printing system himself, or a GUI interface to CUPS.

Instead, he coded... Fetchmail. Which is, to be honest, not that great a piece of code (though it does work). But its configuration UI, Fetchmailconf, is one of the worst UIs I have ever used in any operating system. Ever.

Really?

Posted Feb 28, 2004 8:46 UTC (Sat) by Ross (subscriber, #4065) [Link]

I'm not sure if you are just trolling, but assuming not there are
several differences you seem to have missed.

1) The CUPS developers didn't knock his teeth out. They committed the
horrible act of giving him a free (in both senses) printing system
based on current Internet standards. If that's such a horrible
crime I wouldn't like to see what you'd call the kernel developers.

2) I'm not speaking for the CUPS developers so I have no idea if they
will listen to his rant or not. And either way my point, which was
that the rant was not very constructive, stands.

3) Yes, in fact being polite tends to get you better results because
you avoid pushing the other side to dig in their heels.

4) He's bitching about the wrong people anyway since they didn't create
the stupid GUI he spends the most words ripping on.

5) He shouldn't be one to complain about user interfaces unless he's
willing to clean up his own.

6) If he really wanted to get it fixed he would have submitted a bug
report with plenty of documentation, or, horror of horrors, a patch.
You know, that whole open source collaboration thing?

Really?

Posted Feb 28, 2004 11:55 UTC (Sat) by ren_123 (guest, #19852) [Link]

1) So, if it is free, you have no right to complain about it?

2) If you look at the first five reactions, at the top of this page, I
see some very good comments where UI designers can take advantage of.
Apart from that, his whole describtion of what things went wrong, and
what he did next, is certainly useful for UI designers.

3) yes, being polite does get you better results. However, if the other
party refuses to listen to valid complaints because they're 'not worded
properly' then it becomes a wordgame. You might expect such behavior big
bureaucratic companies, not from open source developers.

4) Ah, he's barking up the wrong tree, so we can ignore him.

5) And before we take his complaint serious, let him fix his own work.
That will teach him.

6) Yeah, raising the bar for complaints is also a good idea.

I'll bet that if you combine the above three points, most users will just
cave in, and pray that the next version has fixed their bugs
automagically. Ever googled for the 'clue train manifesto?'

That whole open source collaboration thing thrives on listening to your
users, and taking them serious.

Really?

Posted Feb 29, 2004 21:28 UTC (Sun) by Ross (subscriber, #4065) [Link]

1) Did I say that? I don't think so. What I said was that releasing
something free is not the same as kicking your users in the mouth. Sure
they can complain. I can complain about your post using invalid debate
tactics can't I. But if I said you're an idiot for the way you designed
the CUPS GUI you would probably not take me seriously. And if I failed to
understand why and said "just because your post was free doesn't mean I
don't have a right to complain about it" you should have the right to
laugh in my face.

2) Those reactions weren't written by ESR were they? I'm talking about
ESR's rant, not posts on LWN.

3) Did I say they weren't listening to complaints? Did I say they
shouldn't listen to complaints? I don't think so. What I am saying is
that it is a natural human tendency to react badly to people who call you
names -- especially when they ignore feedback channels and use their
notoriety to call you names in a public rant.

4) He's barking up the wrong tree so maybe he should bark up the right one.


5) I didn't say we should ignore his complaint because of this but he
probably should check his own work for the flaws he's complaining about
before lobbing such a rant to the public.

6) Once again you mischaracterize what I said. I said that he chose a
crappy way to approach fixing the problem if he really wanted to get it
fixed. I didn't say that user's complaints should be ignored if they are
not "properly worded" or fail to "meet the bar". But if someone like ESR
can't be bothered to complain in a productive way there is little hope for
decent user feedback in general. And it makes me wonder if he really
should be one of our "leaders".

Hmm

Posted Feb 26, 2004 23:07 UTC (Thu) by allesfresser (guest, #216) [Link]

I believe the example configuration file that comes with the CUPS source has these faults as well.

Ah, ok

Posted Feb 26, 2004 23:12 UTC (Thu) by Ross (subscriber, #4065) [Link]

Then nevermind what I said above. That's a pretty confusing disconnect
between documentation and implementation.

I retract my retraction

Posted Feb 27, 2004 18:49 UTC (Fri) by Ross (subscriber, #4065) [Link]

Apparently he was complaining about a GUI that wasn't even written by
the CUPS people. It's hard to criticize them for not being consistent
when they only control one part of the software. ESR should have called
the Red Hat or Fedora people idiots instead.

non-shared default "fault"

Posted Feb 27, 2004 12:38 UTC (Fri) by ringerc (subscriber, #3071) [Link]

I think it worth consideration that if CUPS /did/ share printers by default, it'd be a significant security issue.

That aside, I must confess that I've found printing under Linux and in fact other UNIXes to be a nightmare of badly documented and quirky software. Of course, other OSes don't seem enough better. Windows gains from having vendor-supplied drivers and an interface designed to do simple things easily - but becomes awful as soon as you need to get something not-so-obvious done. MacOS - nice interface, too bad about the backend - especially lpr printing.

non-shared default "fault"

Posted Feb 27, 2004 18:46 UTC (Fri) by iabervon (subscriber, #722) [Link]

It would definitely be a security issue if printers were shared by
default. But it doesn't help much if users don't realize that their
printers aren't shared by default, and why. If sharing your printer is a
security concern, then there needs to be guidance as to how to maintain
configuration with acceptable security while getting what you want to
work.

Copyright © 2012, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds