Log <-

Archive for December, 2005

Backup v0.5

Tuesday, December 27th, 2005

Even though it's unmaintained, I still added a new version of the Backup script: version 0.5.

Changes include:

  • Complete rewrite
  • Much stuff moved to functions
  • Pre, post, pre_each and post_each script support added.

Information harvested by NSA larger than acknowledged

Sunday, December 25th, 2005

From Slashdot:
The New York Times is reporting that the "volume of information harvested from telecommunication data and voice networks, without court-approved warrants, is much larger than the White House has acknowledged." The NSA gained the cooperation of many American telecommunication companies after 9/11 to access streams of communication, both domestic and international, as a part of a presidentially approved program to hunt for evidence of terrorist activity.
What was it again? Oh yeah!

First they came for the communists, and I did not speak out—
because I was not a communist;
Then they came for the socialists, and I did not speak out—
because I was not a socialist;
Then they came for the trade unionists, and I did not speak out—
because I was not a trade unionist;
Then they came for the Jews, and I did not speak out—
because I was not a Jew;
Then they came for me—
and there was no one left to speak out.

Rev. Martin Niemöller

Land of the Free indeed.

MBNetFS

Friday, December 23rd, 2005

MBNetFS 0.3.0.

SMBNetFS is a user-space filesystem for Linux that allows you browse a Samba/Microsoft network much like the network neighborhood in Microsoft Windows.

All I can say is.. Finally!

MySQL storage engine woes

Friday, December 23rd, 2005

Are you suffering from mysteriously failing insert queries due to foreign key constraints errors in MySQL?

Have you noticed spontaneous table storage engine changes from InnoDB to MyISAM, but can't figure out why they're happening?

You may be coming under the influence of the MySQL Command Center which, when altering a table via the Edit table interface, doesn't load the correct table type in the type dropdown, but defaults to MyISAM. This causes each table change you make to revert the table to MyISAM if no foreign key constraints are present in the table. Otherwise, the edit simply fails.

When editing a table, check the 'Table properties' tab before saving your changes. It'll save you a lot of headaches.

Joymaster

Thursday, December 22nd, 2005

I've put v0.1, a preview release, of Joymaster online.

Joymaster lets you control (X) windows using your joystick. Other programs like this already exist, but Joymaster takes a different approach.

First of all, it watches the currently focussed window to see which buttons should perform what actions. This means you can have joystick controls perform different events per window. Secondly, it supports macro's both on the joystick and for controlling windows. For instance, tapping the 'Up' button on your joystick three times can result in an event sequence on a window that loads, changes and saves your document and then exits the application.

I am personally using it to load and save games under various emulators as well as quiting emulators and then controlling a running Midnight Commander to select a different game to play. I also use it to lauch, watch and pause movies using Totem. As you can tell, I'm quite the couch-potato.

OpenOffice needs a draft-mode

Thursday, December 22nd, 2005

There's one thing OpenOffice.org lacks that both Word and WordPerfect have: a draft mode where you don't have to see page breaks and unnecessary layout visuals. To me, this seems like such a basic and important feature. My needs for formatting and fancy features are practically nonexistent–I just want to concentrate on my writing.

OpenOffice Writer does offer a "web layout", but it's just not the same.

I use OpenOffice all the time to dash out letters and so forth, but when I need to concentrate on my writing I always fire up WordPerfect. Lack of a good draft mode is all that's keeping me from using OpenOffice Writer exclusively. I'm sure tons of other writers feel the same way. And I can't imagine implementing this feature would be difficult.

– Schlemphfer (556732), Slashdot

Agreed.

How Google fucked up USENET

Wednesday, December 21st, 2005

How Google fucked up USENET (newsgroups):

Replying to a post somehow doesn't quote the original post, resulting in a lot of posts like this:
Yes! Totally! Good idea
No, that's incorrect. It's actually ment for something else
etc.

In other words.. a total loss of context. EACH message I checked which didn't quote context was posted using Google Newsgroups. Some try to blame this on the users, but I cannot agree. A user interface should make it clean and as easy as possible to do the correct thing. Google newsgroups obviously doesn't do that.

I used to use Google newsgroups for a while until they put version 2 (still Beta, like every other Google 'product' out there) up. They got rid of the tree view, fixed-width fonts, correct replying, etc. Essentially, they made it a blog-style commenting system, and I hate them for it because it also attracts blog-style commenting users. Google's latest Newsgroups version also removes the ability to detect cross-posted messages; a common trolling technique. On top of that, they hide any quoted context that's too long, so that for any ten messages you read, you'll have to perform about six clicks in order to get some context.

(PS: Yes, I'm aware that most of the things I've complained about can be changed using preferences. However, I'm not about to let google, a company that makes all its money from advertising, place cookies on my computer. Next to that, most people don't take the time to change their settings, resulting in more annoyances for me (for instance, in the case of quoting context) )

As far as I'm concerned, this makes google even worse than Microsoft. So far, I've always been able to avoid Microsoft's products if I didn't like them. This, however, I can't ignore because the articles will still show up on USENET.

I hate google.

A DO and a DON'T for OO programming

Saturday, December 17th, 2005

Here's a little DON'T hint for OO programming:

  • Don't use classes to cut up your program.

Here's a little DO hint for OO programming:

  • DO use classes to create small libraries which are all seperate entities that your program uses (and, in essence, consists of)

When writing a program, ask yourself this:

"Can I seperate each individual class from the rest of my program and still use it".

Or just write test cases for each individual class and you'll find out quick enough if your program's structure is okay.

Torvalds: Use KDE

Tuesday, December 13th, 2005

In an e-mail mailinglist discussion on listing, Linus Torvalds wrote the following:

> Frederic told that the options from the PPD file are
> intentionally mot listed in the printing dialog, the
> usability team of GNOME was against listing these options.
> They clutter the dialog and can be more confusing than
> useful to the user.

I personally just encourage people to switch to KDE.

This "users are idiots, and are confused by functionality"
mentality of Gnome is a disease. If you think your users
are idiots, only idiots will use it. I don't use Gnome,
because in striving to be simple, it has long since reached
the point where it simply doesn't do what I need it to do.
In another thread on the same list, he writes:

> That's definitely not a point of view of the GNOME Project
> – we're focused on making Free Software appropriate for
> users who are smart (we don't talk about 'dumb users'), but
> just don't care about computing technology. We're just like
> every other Free Software project – fixing stuff requires
> the work and attention of people who care about the problem
> at hand.

No. I've talked to people, and often your "fixes" are
actually removing capabilities that you had, because they
were "too confusing to the user".

That's _not_ like any other open source project I know
about. Gnome seems to be developed by interface nazis,
where consistently the excuse for not doign something is
not "it's too complicated to do", but "it would confuse
users".

……

The reason I don't use Gnome: every single other window
manager I know of is very powerfully extensible, where you
can switch actions to different mouse buttons.

Obiously, people will all be on Torvald's ass for making these statements, but I happen to think he's right.

Usability does not equal few options

For some reason the so called usability experts have gotten it into their heads that if there are too many options, people will get confused, so there should be as few as possible. This does NOT make a program usable. Take for instance your regular Notepad editor versus Vim. Notepad has few options, vim has a gazillion. Notepad is completely unusable for me as a programmer, because it lacks an enormous ammount of options. If you'd ask which one is easier to learn then I'd have to say Notepad. That does not, however, make it the more usable one.

Ofcourse, Notepad is not designed for programming, but the same goes for just about any IDE I've ever seen. They're all too limiting. You may think the IDE you're working with is pretty nifty because it automatically pops up a little dialog that reminds you what parameters you need to enter when constructing a 'for' loop, but it's all nothing compared to the power of Vim. Does your IDE allow you to quickly append any character (a comma or semi-colon) to a bunch of selected lines? Does it allow you to align all the equal signs on a couple of lines? These are all basic programmer operations which are missing from most IDE's I've seen.

The Dump-People myth

For some reason there's also this prevailing myth that 'dumb' people will be able to use the computer better if the interface is more consistent, has less options, has big nice icons and uses metaphors from real-life (i.e. desktops, folders, etc). I don't know where they got this idea from, but it's not from actual field testing results. 'Dumb' people will not understand current computer interfaces, period. Consistancy may be able to help people who have already grasped a certain aspect of a user interface, but it will never help those that never understood it in the first place.

I've heard countless stories of people how couldn't even grasp the idea of minimizing windows or the taskbar in Windows. They simply move (or even close) all the top windows out of the way in order to reach the window they wanted. Removing all but the most basic options from a computer is the only way to make it easier for people to use the computer. But guess what? That would make a computer virtually obsolete.

'Dumb' people cannot be taught how to use the computer; it's of no use trimming down user interfaces so that they can.

(By the way, I'm constantly quoting 'dumb' because these people are not actually stupid or anything. If you'd put einstein behind a modern computer he'd probably have a pretty hard time understanding all the gibirish on the screen)

Losing the Unix way

Modern interfaces are becoming less and less powerful and less and less usable. Let me tell you two little stories:

When I first got my new laptop I wanted to get the little multimedia buttons at the top working. In Gnome's Keyboard preferences program I found a way to make a certain button adjust the volume. When I pressed the volume-up button Gnome adjusted the volume on the laptop speaker, but not that of the headphone jack. There was no way I could tell Gnome that it should adjust the PCM mixer instead of the one it was adjusting now. I had to write my own little python program that monitored the keys I pressed and then executed an ancient console tool called aumix that COULD adjust any volume mixer I told it too. If Gnome had listed an option to change that preference, I'd have saved a lot of time and
effort.

The other day I was writing a script and I wanted it to open an e-mail in Thunderbird (my e-mail client). After some time I found out Thunderbird simply cannot be told to open a particular e-mail message. Now I use mutt to open those e-mails. Mutt is another ancient console program first written years ago.

Conclusion: Programs written years ago that only have a command-line interface are doing a better job than programs written recently. It's pathetic. We're gradually losing the Unix way of writing programs: Do one thing and do it good; Coupling programs together using their standard input and standard output; Be liberal in what you accept and sparse in what you put out; Keep back-ends and interfaces seperate; etc.

I've been laughed at by Windows administrator when they saw me working on the Unix commandline. "So that's Unix huh? Looks like DOS! HAHAHA". Yeah, very funny dipshits. Funny until someone asks you and me to both produce statistics on the webserver logfile between the two dates. Then you'll have to go and find some idiotic NT//IIS log datamining tool while I'll be done in 4 minutes by doing some grepping and cutting on the commandline. Less options does not mean it's easier to perform your tasks and usability does not mean easy-to-learn!

The ideal OS

For me, the ideal OS / User Interface can be defined as: "Let's you do the task at hand as fast and as good as possible".

When we start removing options from user interfaces because it might confuse those people that will probably never even use those interfaces at all, we're not enhancing the OS or interface. We're just making it harder to do the job for all other people.

The ideal OS for 'dumb' people shows a menu with four options: Chat, Photo's, Surf the web, E-mail. That's it. Anything more is too complicated for most casual users. Does that mean all other stuff should be removed?

Gnome usability-team

The Gnome usability team, along with most of the rest of the so called usability experts, need to understand some things better. They're trying to make a compromise in the user interface between power users and 'dumb' users by making an interface that's somewhere in the middle. This is not a good idea, in my oppinion.

For instance: The discussion mentioned at the top of this post is about PPD options in the printing dialog of Gnome. These options aren't listed because they might scare 'dumb' people. Let me tell you something: 'Dumb' people do not look at the printing dialog. 'Dumb' people press the print button and expect stuff to come out of the printer. If they want stuff printed twice, they won't enter '2' in some inputbox: they'll press the button twice. If they only want pages 1 through 5, they won't enter '1-5' in some inputbox: they'll just print the whole damned thing and throw away anything they don't need. People are like that. That doesn't mean the whole Printing dialog needs to be removed just because people don't use it, does it?

Look, put all the fearful, complicated, little buttons and options in some 'Expert' tab for all I care. Just don't remove them okay. Oh, and add command-line options and non-interactive modes to programs because graphical user interfaces suck.

When it comes to Gnome, I'm with Torvalds on this one, but these problems are not limited to Gnome. I haven't used KDE in a long, long time, but the last time I did, it wasn't any better than Gnome at all when it comes to usability.

OfficialWire: Rectify the article?

Tuesday, December 13th, 2005

I e-mailed the following to OfficialWire to see if they hold themselves accountable for faulty news reporting:

In your news story titled 'It's Your Story…You Tell It Anyway
You Want On Wikipedia' the following claim is made:

"After four months, Seigenthaler was finally able to get Wales
to remove the offending piece from Wikipedia and from the other
online 'resources' that simply copy from Wales' pool of data"

However, there is no indication in the original article written
by Mr. Siegenthaler (titled: 'Truth can be at risk in the world
of the Web') that this was the case. In his article, he writes:

"The 'biography' was posted at 2:29 p. m. May 26. On May 29, one
of Wales volunteers, identified only as SNlyer12 'edited' what
had been written about me only by correcting the misspelling of
the word 'early.' For four months, Wikipedia continued to
present me as a suspected assassin and a 13-year resident of a
Communist regime."

This paragraph seems to be the source for your claim. However,
he does not claim it took four months to remove the information
after he contacted Wikipedia or even found the article. He
claimed the information had been on there for four months total,
not four months after he contacted Wikipedia. There is no
indication in Mr. Siegenthaler's article of when he first
visited and/or contacted Wikipedia.

I could have missed something in the original article, or
perhaps you've contacted Mr. Siegenthaler himself and got the
information for your claim via that route. In any case, please
re-check the claim and verify it's accuracy.

In the case that the claim is found to be wrong, what actions
will be taken by your organisation to correct it?

I wonder how they'll respond.