While we’re on an OpenWRT kick, let me mention that Amazon has the guaranteed flashable WRT-54GL for less than $50 again. I won’t even provide a non-referral link, so as not to raise eyebrows at wordpress.[non-]com[mercial]. I’m sure you could find it.

Possibly the best reason someone who’s not interested in getting to know their network or doing fun things with it is the possibility to use just one OpenWRT/DD-WRT/Tomato device (along with any standard wireless router) to extend internet connectivity to another part of the house without running wires or using shaky (and unencrypted) Power-Over-Ethernet.

On my list of future projects to waste my money on, though, is a set of cheap 802.11n devices, replacing the “G” bridges I’ve had for the last few years. Apparently, that day is near or already here. Of course “N” speeds only matter if you’re concerned with moving data between machines in the home. For a normal residence, any available residential networking technology is faster (in most cases orders of magnitude faster) than even the best of the universally crappy available residential internet connections, and so the bottleneck will always be at the WAN side (especially for upload.) Of course these days, most small businesses are actually paying more for even crappier Internet service. But I digress…

On a last OpenWRT note, if I had known about this contest in time, I definitely would have put together a team and devoted a few months. Maybe it really does pay to read Slashdot.

OpenWRT Kamikaze 8.09 processes /etc/config/dhcp and appends the settings specified there as command line options when running dnsmasq. To properly configure dnsmasq to point to your LTSP server you will both need to specify options in this file, and modify the dnsmasq init script for these new settings. (Note that you can also—right now—add an /etc/dnsmasq.conf with these options in standard syntax. But that’s not the way OpenWRT is heading, and there’s inconsistency across the various packages as to the use and naming of non-uci config files.)

The following worked for me with Debian Lenny running LTSP 5.1.10 installed fromt the repositories, other distros may require different paths:

In /etc/config/dhcp, add in the “config dnsmasq” section:

option dhcp_boot ltsp/i386/pxelinux.0,$LTSP_SERVER_HOSTNAME,$LTSP_SERVER_IP
option dhcp_option 17,'$LTSP_SERVER_IP:/opt/ltsp/i386/'

The first path is relative to /var/lib/tftpboot, the second is the ltsp chroot directory which should be in /etc/exports. The second option is equivalent to the root-path option offered by dhcpd (AFAIU) and solves the “need path” error. Apparently this is not necessary when the DHCP and LTSP servers are the same.

Etherboot clients may require different path information. And if you have a mixed set of clients, you can use the dhcp-vendorclass syntax demonstrated here (the paths in the dnsmasq section of this document are out of date, though they have been fixed in the rest of the page.)

In /etc/init.d/dnsmasq add in the “dnsmasq()” function:

append_parm "$cfg" "dhcp_boot" "--dhcp-boot"
append_parm "$cfg" "dhcp_option" "--dhcp-option"


Run:
# /etc/init.d/dnsmasq restart

Then boot one of your thin clients, hopefully to an LDM login screen.

(Credit where credit is due: This post on general PXE booting with Kamikaze got me halfway to getting this to work.)

OpenWRT is one of those great projects which suffers for lack of documentation. Even worse, there have been three generations of configuration methods, and the existing documentation and user-written howtos can often be vague about which version they refer to. Some commands are future compatible (meaning they will work on later versions of the firmware) and some are not. Some may work, but in less than optimal ways and leave you with a mess to maintain in the long run.

(A similar problem exists with Xorg where many, many users who aren’t and don’t want to be X gurus but who know a few tried and true xorg.conf hacks have gone into their systems over the last year only to discover they no longer have an xorg.conf. Now, maybe you can just add a file to replace the non-functioning background magic, but you’ll be left wondering what you’re giving up, what might break down the line, and so on. And you probably wont find a simple answer for a modern system yet. You’ll have to wade through mountains of mailing list and forum postings, and reconstruct the workings of the new style X server in your own mind, which is exactly what you wanted to avoid.)

So, for those who don’t pay much attention to these sorts of goings on in OpenWRT land (how often do most of us reconfigure our networks?), here’s a set of rules of thumb for recognizing the version of OpenWRT any given online tutorial applies to:

• If you will be changing NVRAM settings, it’s for the old WhiteRussian release. (You should probably upgrade if you’re still using this.)
• If you are editing config files which resemble those of a normal Linux system in /etc, it’s for Kamikaze 7.x series. Many of these config files will still be read in the current 8.09.x series if present, but don’t necessarily count on it. You should be able to figure out the launch process for each package by reading the init scripts.
• If you are editing files in /etc/config or using uci, this is for Kamikaze 8.09 or later. In this series the LuCi web interface is also on by default, if that’s your thing.* Occasionally something shows up in the forums which uses uci directives which don’t seem to exist in stable. Also, you may find yourself adding options which the init scripts do not yet process, so you will have to edit them in.

OpenWRT is indispensable, and like much great software, those using it would rather keep using and improving it than document it and clean up the existing mess of documentation. Hopefully this helps with a little of that weeding for the new user.

* You can remove the LuCi web interface from OpenWRT Kamikaze by killing the process and running:
opkg remove -recursive luci-*

Morning Edition is talking today about the usual troubles the census goes through to get everyone counted. Apparently just about everyone in NYC belongs to one of those groups least likely to return a census form: young and single, immigrant, or African American. Apparently Hispanics are an exception to the immigrant trend for whatever reason. One of the suggested reasons for immigrants not returning census forms is their being unaware of what a census is. Is that really possible? While I’m sure that some of the states which haven’t had a functioning government in decades might miss out on the fun, I’m not personally aware of any country without a census.

Actually, I assume that in most countries it’s more important than it is here. Much of the developing world uses some form of representational quotas to fill their legislative bodies and apportion cabinet positions. The result of that could be that it’s more contentious or that it’s more corrupt, but either way everyone has a stake in making the census prominent in the national consciousness.

I was in Turkey for census day in 1997. Not having any advanced warning, that was the day we had chosen to get back overland from Trabzon to Istanbul, via Giresun. This was made nearly impossible by the fact that every 50 yards or so across the entire country there was someone dressed like this demanding papers of everyone on the street:

Apparently no one is allowed to leave their homes on census day unless they have advanced parole. The dolmuşes  (shared taxis—yeah, I’m sure that plural should be made with a ‘lu’, but you get the idea) were getting stopped constantly and everyone had to show their permission slips. It was only in Giresun that we finally found an English paper that explained what was going on. (No one on the Black Sea coast seems to speak English and either I couldn’t figure out what Volkszählung meant or they were using something else to describe it in German—everyone in Turkey speaks German, while I only pretend to—or they were using some other phrase to describe it. I’m guessing the latter because the word seems pretty obvious 12 years later and in writing.)

I just wonder what sort of effect that type of experience has on people—would they be more or less likely to return a US census form? Would it have to do with fear of the census or the need for serious encouragement to actually bother with it?

I love this structure:

<dl> <dt>Animals</dt> <dd>
<ul>
<li>Cat</li>
<li>Dog</li>
<li>Horse</li>
<li>Cow</li>
</ul>
</dd> </dl>

- Kevin Marks on [whatwg] List captions

It seems more correct than the alternative he proposes (with each of the li’s here as dd’s) in that the list is an ostensive definition, whereas each of the items is a mere example. Practically, it saves me from having to deal with some lousy CSS on nested lists. And it does feel more semantic-ish than the nested list (for this case) anyway, and certainly better than the usual headings or paragraphs inside lists.

As Fears Grow Over Pakistani President, U.S. Woos Rival

This paragraph is the only real background on NS in the article, and could at best be called counterproductive or misleading:

Mr. Sharif, 59, represents the Pakistan Muslim League-N, a coalition that includes a number of Islamist groups. He was prime minister twice during the 1990s, and received hero status in Pakistan for ordering nuclear weapons tests in 1998.

(Either my headache is more severe than I thought or they changed that in the last twenty minutes from “Muslim League of Pakistan” to PML (N)—there goes 25% of my rant!)

Anyway PML (N) is not a coalition, though presumably they are referring to the APDM which was made up of every party—Islamist, secularist, feudalist, opportunist—which actually wanted Musharraf out at the end of his reign.

His ties to Islamists are nothing but pure political calculation. The PPP and PML have each taken their turns flirting with the religious parties or the military or the West or the Saudis, and neither party has ever had even the slightest hint of an ideology beyond their own political survival, the humiliation of their rivals, and the advancement of the material well being of every hoodlum willing to rob the country blind under their banner.

These two off the cuff aspersions posing as biographical notes are the sort of media fearmongering I don’t get. If this were typical NYT South Asia reporting (read the Hindu, call up Sonia, send out your summary and call it a day) I might understand. But even though there was a month or so last year where Benazir could do no wrong, this article is pretty harsh on the current state of the PPP and President 10%. So if you’re going to bring this up (rather than saying that NS and his party represent the Punjabi feudal elite’s interests or something else vaguely social sciencey) you could at least do the balancing thing:

Nawaz is currently in a pro-democracy coalition (though not an electoral coalition) with an inter-sectarian Islamist party whose factions all hate each other but hate the secular state more (and whose biggest factions didn’t even participate in the last election) along with some crazy almost-Commie secularist Pathans and a crazier athlete turned megalomaniac politician. The ruling party under its recently martyred heiress oversaw the creation of the Taliban. Nawaz conducted nuclear tests in a deliberate provocation of India. Benazir smuggled nuclear plans to North Korea in her handbag.

I’m just saying, if you want to scare us, go all the way. Americans are constitutionally predisposed to relate to scenarios of total depravity.* Fulfill that need.

(The exact same two points—sharia and nukes—were also used as the entire background in a nearly identical article a month ago.)

As for the news itself, if it’s just a case of America deciding it needs a new dance partner, then it is tragic. If it’s us saying to Zardari that he needs to work with all centers of influence and power, it’s great. The article suggested it’s a little of each.

Personally, I would appeal to both of them—and to Imran Khan and the military leaders and the religious parties and every 5th grade student in the country—by saying that the balance of their legacy is extremely negative, and that if they can’t just once put aside their selfish, desperate pursuit of every last pai left to be pocketed they are going to lose their country for good and it will be their fault and history will remember.

* If you find that sentence insulting, you probably didn’t understand it. If you find anything else in the article insulting, you probably belong to a Pakistani political party and were meant to.

So the last post caught Habib Haddad of Yamli’s attention. Which brought his twitter feed to my attention. Which led me to this article from Flip Media. Which led me to realize that in addition to two other yamli competitors, Google already has their own ta3reeb. I still love yamli – and their search component is as great as the keyboard aspect. Still I’d love to spark some competition between them for inclusion of characters from other Arabic-based scripts and some of the presentation forms.

Now, about the competitors:

1. Google’s ta3reeb: I guess the existence of this makes it more likely to be integrated everywhere, on the heels of (maybe even within, like I dreamed in the last post) the Indic scripts gadget. It also makes it less likely that Google will acquire Yamli. Insha’ Allah, those guys will get rich anyhow.
   At present, though, note that this means I don’t have to give my contact list to anyone who doesn’t already have it. The URL for adding this gadget to GMail is http://www.google.com/ig/modules/arabic_transliteration.xml.
   Ta3reeb responds more quickly than yamli but in a choppy and not altogether useful way. If you’ve used word completion you can understand how over-responsiveness is not always conducive to efficiency.
   Now, why haven’t Google added this in the Language options on GMail? There is—as you might expect—a Google Group where you can go and pester them about all of your hopes and dreams.
2. Eiktub has a Windows Notepad-like editor you can download, in addition to the on-screen keyboard. The behavior is a little bit different than Yamli or ta3reeb – there’s no drop down suggesting completions. I did not test the app, though it looks nice for what it is. It could stand as a Windows substitute for Katoob. Really the only app I switch to Windows for is InPage. So this might be redundant. I will try it, eventually, though. Tashkil in the web interface is a nice touch, though it seems to eat the fathas on lin vowels. They lose points on the “openness” scale, though.
3. The last of these is Onkush. This is really the least effective – or rather the only ineffective – of the online dynamic keyboards. My “hello world” for testing these is to type “salam.” Onkush is the only one that defaulted to سالم. I don’t really know how the weighting is done in the others, but this is a bit of a failure. The interface also takes a long time to do the analysis and conversion, and seems to freeze with too many backspaces. It is interesting that they have pages for different types of search like Google. Though I’m not sure about their default set of pics on the image search page.
   Onkush was the only one of these where I started to fire up Firebug and look at how it was working. I knew when I started that it would crash my browser, and it did.

Now if any of these were Open Source projects, I would put up a bounty for expanding the character set, or do it myself. I’m sure an entire free replacement might be doable as well, to which I would probably be willing to contribute my time, so-so JavaScript skills, or maybe even money. Anyone interested? Does such a project already exist? I always seem to fall behind on these sorts of developments and have to catch up once a year or so. (As evidenced by the fact that either I didn’t know about ta3reeb or just forgot.)

Yamli is an Arabic frontend to Google (I believe I’ve mentioned it before) which deserves to win all kinds of awards. (Click to skip over verbosity to the howto.) They use a Google suggest-like interface to convert phonetically typed words and phrases to Arabic script. It’s perfect for on the fly typing of short passages on machines where setting up an Arabic keyboard mapping or switching is just not worth it. It is very similar in operation to the Indic scripts gadget which has slowly propagated across Google’s properties, and was recently added to the Gmail editor.

Yamli also offers a stand alone “editor”—really a bigger text box with the ability to print what you type or save it as an RTF document. ([RW]TF?!?! I’m sure there’s a reason for that, but there’s a thousand arguments I can think of against it.) On top of that, there’s an embeddable version (quite customizable), and they’ve pre-created widgets for iGoogle and Facebook.

I’ve become quite addicted to yamli over the last few months, and these are a few things I’ve thought about that would be great to see done:

• Obviously, the first thing I would want to see is it extended to at least the Farsi and Urdu character sets. Sindhi and Pashto would probably just overwhelm the options, so I’m comfortable discriminating against them. I’m Kidding! (Those are two language groups you probably shouldn’t piss off…) These are all in Presentation-Forms-A anyway. (Burushaski and Divehi, on the other hand…) Maybe there could be embedding options to turn off extended character sets. And automatic conversion for things like: ﷺ.
• I would love it if the search bar version worked like the version on the site. (This should be possible, right, considering the way Google and Wikipedia suggest search plugins work?)
• It would be great if in addition to suggesting character conversions it hooked into Google suggest and suggested search term completions. That might be overload though.
• Google could buy out yamli or both parties could open source their work, and we could have a combined widget for Indic and Arabic-based scripts. On top of this, add whatever else is out there for Hebrew, Cyrillic, Greek, and CJK scripts. Allow some sort of configuration to enable and disable language sets. Make it an option throughout Google and embeddable everywhere. (Why not just make this a default behavior of text inputs in html5? I’d even settle for a unicode-typing-text element. I’m sure the working groups are all reading this.)
• I’d really like to have this same unified phonetic typing magic in a text editor or as something OS-wide, or at least DE-wide. This would eliminate a lot of pain and wasted time. Take me as a case study. I type text daily in the following scripts in descending order: English, Urdu, Farsi, Arabic, Roman script with diacritics, Hindi, Telugu. On my main keyboard I have Arabic key stickers. I have memorized the Urdu phonetic keyboard from CRULP . And Indic keyboards are mostly phonetic, but I do have to do a lot of backtracking to get things right. There are a few diacritic shortcuts I know, and the rest I cut and paste or use an on-screen keyboard. I know there are plenty of people who type in a number of scripts on a daily basis. We could all use a tool like this. I’ve transitioned from mostly Emacs to mostly Vim, but I would switch in a heartbeat to whoever enabled this first. In an editor, the Google suggest integration could be replaced with actual word completion based either on a dictionary or the rest of the document, similar to other word completion interfaces.

I’m pretty sure it should be possible to write a user script which yamlifies the Gmail editor. This would be a second-best in my book to the unified version, but most Arabic typing types probably could care less about Indic scripts. So the next best thing until someone does that is to have a yamli widget in Gmail itself. This, it turns out, is within reach of even the laziest among us. There are three simple steps:

1. Go to your Gmail settings and enable “Labs.”
2. Go to the Labs tab which is now in your Settings, scroll down to the bottom (enabling anything else you like along the way) until you find “Add any Gadget by URL.” Enable this and save changes.
3. You now have a “Gadgets” tab in addition to the “Labs” tab. Go to this new tab and add the URL for the yamli iGoogle Gadget: http://www.yamli.com/google_gadget/

If you are using NoScript (you are using NoScript aren’t you?) you will need to enable $crazylonghash-opensocial.googleusercontent.com, and clicking in the yamli box without doing so will result in clickjacking warnings. I suppose this also means that yamli’s developers have access to your “friend’s list”—i.e. your Gmail contact list. Which might make you wonder if it’s worth it. But who am I to make that decision for you, I’m just the guy who wants to rewrite w3c specifications for my own obscure purposes.

UPDATE: You can have your dynamarabic keyboard magic and your privacy too! (Or what little is left after Big Brother Goo is done with you.)

There’s an environment variable HISTFILE which determines where ksh saves your command history. By default, if HISTFILE is not set, ksh will save to ~/.sh_history. OpenBSD does not, however, include this default – so your command history is lost at the end of each session. That may be what you want, but if not, you can just add a line like this to ~/.profile:
export HISTFILE=~/.sh_history
Since ksh doesn’t include source you can use . ~/.profile to load your changes.