Older blog entries for chalst (starting at number 284)

Turning circle

Cited with little context from a post of Russ Allbery, regarding the decision to create Canonical:
Debian was then, and is even more today, a huge ocean liner of a project with a very bad turning radius.

Incidentally, are redi and badvogato the only members who are still posting unsyndicated diary entries as often as most months? Advogato's circle for turning out of the way of spammers can also be considered rather bad. I have a suggestion for a patch that should be simple enough to be feasibly applied that should fix the problem. I'll post an article shortly.

Shameless

From the Advogato page at knowem.com:

Members of Advogato are commonly tagged with:

social media, used car dealer, seo, dealership, Honda, Honda Certified, Honda Civic, Honda Dealer, Honda Vehicles, Search Engine Marketing, Search Engine Optimization, Social Networking, Accord, advertising, artists, Auto, automobiles, automotive, bjj for kids in jacksonville, Car, car dealer check, car dealer reviews, car dealers, Carpet, cars, Ceramic Flooring, chat, community, Dealer, deals


Click on those links to find lists of people who advertise that they have been creating spam accounts. So far I haven't found any that still exist.
15 Sep 2013 (updated 15 Sep 2013 at 15:01 UTC) »
Where is the mod_virgule TO-DO list?

The project page for mod_virgule gives as its project TO-DO list a link to
  http://rainwaterreptileranch.org/steve/sw/mod_virgule/todo.html
but that is a broken link.

Is there a most-current version of the project's TO DOs?

8 Aug 2013 (updated 8 Aug 2013 at 09:16 UTC) »
Long time

Has it really been over 21 months since I last posted to recentlog? I've been more active with my account at G+.

I'll crosspost something I wrote there: a recommendation for cabal-dev as a way of making Haskell package management sane:


Cabal is a package manager that by default compiles packages either to a system database (if invoked as root/admin) or a per-user database. Since Haskell code is very often fussy about exact version numbers, and because Cabal offers essentially no way to uninstall packages, this is very painful. cabal-dev is a front-end to Cabal that gives you per-directory repositories: suddenly conflicts go away, and you can remove packages by deleting the directory's repository and reinstalling everything again. It's a kludge, but it is very helpful.
28 Oct 2011 (updated 28 Oct 2011 at 12:06 UTC) »
Feeding the backstabbers
mjg59 wrote about how his experience of talking about the Linux experience to the Linux community differed from someone else.

He cited an HN thread, Female FOSS dev quits tech industry due to harassment, and in particular this post from jxcole: I was going to start a rant about how this behavior is encouraged by the macho men online, but this was just one guy harassing her. "Due to harassment" reads as due to harassment from the community, but she gave in to one idiot. She let him win.

I answered as follows:
Well, I kind of agree with this in part, but it is often trotted out as a reason why it's the victim's fault. Let's see what's badly wrong with it by looking at things another way.

Standing up for what is right is not something that we weaker mortals should leave to the heroes. We have a breaking point, a point above which we can't take it. That does not mean we should not take risks to do what we find right. You should not be "disappointed" with the victim, but supportive, and we should be encouraging others to take these risks to do what is right as well. The more of us there are, the we achieve.


On a more general point, be aware that if you regard yourself as supportive of women in free software and on their side, but you are vocally undermining of said women when they face this sort of shit, be aware that you are a fair-weather friend and some sort of a backstabber.

Since I'm on the subject, let me plug Skud's long post, On being harassed: a little GF history and some current events.

Despite the event that lies behind this post, I'm more optimistic than I was a year or so ago that the free software community can sort itself out on this one, since the fair-weather friends do seem to sometimes be a bit more aware that there might be something wrong with their behaviour. Skud deserves special praise for her efforts that have made a positive difference, as does Matthew Garrett. who I suspect will not find himself unable to continue with free software.

C migration strategy
wingo asked, after summarising the safety issues with C:

I still write C. I work on implementations of safe languages -- languages that don't have the same kinds of fundamental vulnerabilities that C and C++ have. Eventually the amount of C in the world will stop growing, and decline as pieces that are now written in C will be written in Python, in JavaScript, in Guile: in short, in languages that don't launch the missiles when you try to write beyond the end of an array.

C has had a great run; we should celebrate it. But its time has passed. What is your migration strategy? How are you going to stop writing C?


I think the most conservative strategy is the best. I was impressed by Cyclone. I'm not sure the language is quite right for the job, and development seems to have halted some time back, but there have been alternatives proposed. I think Cyclone has shown that C can be modestly extended in such a way that we can statically for absence of undefined behaviours and many other desirable safety and correctness properties without nuisance or violation of the spirit of C. I hope that an obviously right safeness extension to C gets proposed and acquires momentum.

I actually had a dream about this, where I was reading a paper about an extension to Lua that allowed Cyclone+Lua programs to have whole-system correctness guarantees. I don't often remember my dreams, funnily enough.

Recentlog
nutella writes, of the site's recent glitches: Okay, now thanks to bad RSS feeds the Advogato front page is really messed up. The main target seems to have been the "Recent People Joining" list.

Bad RSS feeds? Really? I guess it's possible, but I assumed that StevenRainwater was backing out code that was not working, or has safety issues arising from the recent libxml2 migration.

6 Sep 2011 (updated 7 Sep 2011 at 08:22 UTC) »
Ex commands under Emacs
I've only recently noticed something rather useful for me: viper-mode's support for the ex command set, implemented in the viper-ex package as a part of the Viper vi-emulation package, is mostly orthogonal to the rest of Viper, meaning the Viper implementation of ex can be used without activating Viper mode. Hence you can bind the viper-ex function to a key in any Emacs mode and use commands like .,$s/wrong/right/g there. It is quite straightforward to use this to support vi-like macros, which are sometimes quite a bit briefer than their Elisp equivalents.

To get started, you need to import the viper-ex package, which defines the viper-ex interactive command, and which has some dependencies on the rest of Viper, ones which could be pruned without much difficulty. The ex commands are more pleasant with line numbering. The following Elisp will configure this, with viper-ex bound to C-:


(require 'viper-ex)
(require 'viper-keym)
(require 'viper-cmd)
(require 'linum)
(define-key global-map (kbd "C-:") 'viper-ex)
(global-linum-mode)

Note that several modes, like Org-mode, clobber C-: and so would need their maps redefined to use this.


Advogato's new libxml2-based text handler

This is new, I think. It seems to have fewer glitches than the old code, so kudos to StevenRainwater for getting this done.

4 Jul 2011 (updated 6 Sep 2011 at 12:36 UTC) »
Recentlog
ncm writes in favour of τ against π: It's silly to memorize an absurd number of digits of pi and then be obliged to double them before they are useful.

I've no issue with having both constants about, using π or τ whichever is simplest, but I find that π feels more fundamental to me. A couple of considerations:

  1. The area of the unit circle is π. This seems as close to fundamental a fact of trigonometry as you can get. Perhaps trigonometry differs from folk geometry in finding the radius of a circle more fundamental than the diameter?
  2. Quite a lot of the Tau Manifesto is devoted to how much nicer things are if we express radians as lying between [0,τ] rather than [0,2π]. Well, radians are an equivalence class over all the whole continuum, and choosing to express radians over a positive interval to me strikes of failing to appreciate the symmetry of geometry. I find [-π,+π] more natural than [-τ/2,+τ/2], and I like having π being the opposite angle to 0. I do, I confess, view expressing right angles as τ/4 rather than π/2 more intuitive, since right angles "quarter" the plane.

I'm quite happy for people to prefer working with τ rather than π, but I would be unhappy with a calculator than had a τ key and no π key.

275 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!