Unhappy note, but an intersting fellow from the circle I occasionally run in passed away recently. Not much to say, the freeman article pretty well covers it. About the best you'll find of marty on the web, one of the more interesting tracks, although tracking down a full version of a Celia album he was in might be wise. Joys of a weekly house 'jam' session in our apt and bar conversion (mandolin, banjo, guitar, bass, occasionally keyboard and dulcimer), always was entertaining (and educational) to see a master at work with their instrument.
So, RIP marty :-/
Whee, another confcache bugfix release. Version 0.3.3 is up in usual location. Adding it to portage shortly, since the bugfixes against 0.3 have been pretty minor, plus 0.3.x will sit for a bit while some refactoring occurs to better handle doing tricks to avoid invalidation of the cache.
Many thanks to Ed Catmur, Christian Lemke, and Diego 'Flameeyes' Pettenò for feedback and bug reports.
Confcache, version 0.3. Fixed up a bunch of ass biter issues in dealing with the sandbox when trying to start it up (versus confcache being called within a sandbox env); pretty confident it now works fine under userpriv, and standalone as of version 0.2
Wound up having to introduce a lovely little bit of twisty code- certain sandbox vars defined prior to starting the sandbox result in the sandbox not appending the usual sane defaults (allowing write to /dev/tty for example). The solution? Well, we shift those vars to the side, spawn an indirection script that appends them to the sandbox default, then execs the actual configure call.
Kind of nuts, but does the trick.
Aside from the quicky version 0.2 release, 0.3 was released a while after. Closed up some nasty gaps in staleness detection. At this point, I don't know any way to trip up the detection code- still get some occasional failures (mono for example), but the failures so far haven't been confcache's fault, issue has been that the package doesn't properly support --cache (which confcache uses).
Aside from that, rewrote the portage integration patch- back out the original if you have it, and raid the v3 from the usual place.
Addressing a couple of questions-
Why was configure complaining about stale build_alias/host_alias?
Bug, namely. Version 0.3 now digs through the args passed to configure, and (should) properly handle staleness detection for that facet
of configure caches.
Why isn't configure caching all configure tests?
Because the configure script has to be written to cache results. A
lot of default checks are cached, but a lot of package specific checks aren't written to cache (yell at upstream in other words).
Why isn't confcache working for ebuild xyz?
The current integration is active only for econf (it's the only place I can force it).
You can do a nasty little trick however to have confcache step in in the meantime. Add
function ./configure() { $CONFCACHE ./configure; }to your /etc/portage/bashrc. Shouldn't break stuff, but if it does it will be obvious- additionally, don't blame me. It's an evil hack, you've been duly warned :) .
Why is the global cache invalidated every run when I have cpufreq enabled?
Cpufreq changes the proc frequency (good thing), this info is exposed in /proc/cpuinfo, a file configure scripts check. The md5 changes
every time the proc changes frequency, thus invalidating the cache each time (bad thing). I'm aware of it, and intend to address this in
0.4.
Offhand, probably end of the weekend for 0.4- need to introduce a framework for registering triggers to override the normal md5 checks for certain files, so it'll require a bit of expansion to confcache.
Got a little ticked off at the runtime cost of configure calls for package updates, so I wound up rewriting confcache from the ebd portage line as a stand alone tool (score, I can use it locally for development), and split a patch for portage stable integrating it as FEATURES="confcache". Ebuild, and portage patch are available in my devspace.
Rough example of the improvement run time wise, is a reduction of php's configure call from 80s to 25s. About what ebd was getting, if I recall correctly.
For those unfamiliar with what this feature/code is actually doing- autoconf scripts have the option to cache settings in a file, and reuse that file next time around. Save the result of checks/tests, basically, which can result in pretty massive speed ups sometimes.
The problem is that autoconf doesn't offer a way to quickly/cleanly verify the cache- for portage, this is a major issue since a glibc update should most likely invalidate the cache; you don't want configure tests using old results that are no longer correct.
Enter the sandbox. :)
Via the sandbox, we can track what the configure script accesses for it's tests. Store a list of the files checked, and their md5 checksums, and you have a way to check if the cache is stale or not. It's actually a bit more complicated (need to track env vars also), but that's the basic jist of it.
Either way, looking for testers and feedback- the gain from it should be obvious, so I'll avoid the usual pleading to get people to test stuff ;)
For the non-gentoo folk, well, grab our sandbox, install it, and install confcache and you will be able to use it to manage a global cache for configure calls. Still useful, although y'all will have a few more hoops to jump through ;)
Wow, ain't that fun. Follow up to the earlier mayhem of my name showing up elsewhere, another Brian Harring aparently wrote those articles and I've got vague threats of a lawsuit in my mbox. Rather odd, stating that the material was in fact stolen from DoD systems, stating I aparently 'hack' websites-
Brian: This could be considered adverse interference with a contract. At this point in time, I suggest you consult an attorney. I would like to point out that your recent emails to me indicate clearly that you know us, have sent us material before, hack websites and do not like the government. Things like this are better left unsaid but in your case, discretion is not the better part of valor and I will send your emails, and my responses to them, to, among other people, my publisher.Copy of the email is here. Basically, somehow I'm being viewed as interfering in a contract that thus far has been represented as me signing off on with the publisher emailing me (somewhat on the attack I might add)... it's bloody insane.
Fun part is that the publisher sent an email questioning whether "did you or did you not hack into the DoD to get this report?". Email copy available here.
For anyone unable to read between the lines, no, I've not done anything of the sort, continuing, I have no affiliation with tbrnews aside from knowing Pete (author) and Aaron Johnson (his roommate). I'd expect Aaron has little to do with this at this point, although Pete sure as hell isn't exempted out of it. Fun fun.
Aside from that, aparently there is now a brianharring yahoo email address- wonder what the creation date was for that account...
So in short, either I can continue to regale people with tales of this insanity, or just hand the whole stupid mess off to authorities/legal parties and let them deal with it. Rather insane, but only course that exists that, so guess that's the route to go.
Final comment, people kind of suck.
Clarification, made a few updates to this posting, mainly since it seems like this has become a bit of a character attack.
Ain't it fun kiddies, going by publisher's statement, aparently I'm a DoD
cracker, or a
wannabe going by tbrnews. Mildly insane? Story isn't even the same.
Sigh. Any good lawyers in the Milwaukee WI
area would be quite nice to know (not surprising, but factual bits of the attack are wrong), if they're pro bono even better
;)
AUTHORS Note: rather then provide google juice, have shifted the referenced articles into a tarball available here. All links replaced pointing at said tarball. Plus side, the articles are snapshotted so people can't change them.
Recently I started receiving emails thanking me for exposing the 'truth' about the number of dead soldiers in this lovely little Iraq war. Found it a bit odd since I don't recall writing any such expose, but figured there was a collision in namespace, some author popped up with the name "Brian Harring". Thus far I've effectively owned the search results for my name, since no one else with my name seemed to be active in the realm of the internet. Collision I expected to happen at some point with over 6 billion people in the world. So I checked into this other Brian Harring, and was more then a bit pissed off to find out that this author I knew rather well (and no, this isn't "A Scanner Darkly").
Background info, growing up in WI I met a rather eccentric fellow living in Monroe WI during my senior year of High School, named Peter Stahl (Pete for short), pen names of Walter Storch, Gregory Douglas (his son's name), and a dozen others I'm sure. Figure of ill respute, bit off the mark, but interesting to discuss stuff with. After a while, a friend of mine from Juda WI and Pete moved down to Auroroa IL, and I visited on occasion. They were friends- may not have agreed with the crap Pete publishes, but that's his thing. His efforts involved in publishing/writing/whatever are typically pretty much way off the beaten path, but as a general rule of thumb I kept my nose out of it. Commenting on it now, it's mainly sensionalist stories, and that's what I'll leave it at; I have no interest in getting sucked down into questions of veracity of his material, since he's the only one who truly would know, and frankly I'm not sure he cares if others believe it or not.
I take people as they come, judge them by their actions towards me (a moral code pete shared). So a rather odd form of friendship existed, I'd forward moveon.org type material his direction, discuss swords, etc, and wound up meeting a collection of people he works with, some palatable, some not. Interesting folk at least. Time progressed, still talked occasionally, but I wound up bailing on talking to them (and half the people I knew) when I made the move to California in Sept. of '04. Pretty much was intendant on doing something with my life and moving on, seeing what was out there rather then biding my time looking outwards from a dieing WI town. Never held interest in Pete's literary attempts, but by that time that was pretty much his entire life, and my friend (his roommate) had moved on with his life, busy with real world work. Nice way of saying I kind of wrote him off, but neh, it is what it is.
Pete has been running a website named tbrnews.org for around 2 to 3 years now (presumably still)- originally spawned from a group out of California calling themselves the Barnes Review (rather infamous going by googling), although control of the actual dns record was in Pete's hands, not Elizabeth and Willis Carto, so pete went off in his own direction, publishing whatever suited. Looks of things as of late, the site is mainly with an anti-govn. stance (mainly 'King George'/Bush as the target). His recent stories written by one Brian Harring have been drawing attention/readers via claims that the administration is lieing about the actual count of dead soldiers in Iraq, supported by a supposed pdf in the author's possession.
Funny thing, my name suddenly popped up on his site publishing this war crap, and a rather pathetic computer articles. Seriously- computer codes? It is called code, program instructions, scripts, libraries, etc. The scary thing actually is doing a sitewide google search for my name.
Why was my name used by a supposed friend? I have a web presence (google juice), and unlike all of Pete's other pseudonym's, I actually have at least the appearance of a credible reputation- extra distance can be extracted from my pre-existing reputation rather then having to build up credibility/reputation via another pseudonym. Quite usable name to ride on the coat tails of when you're attempting to avoid the loons who like to debate stuff. Longer you can coast without people screaming, the more you can stretch things to try and pull in new readers, or potentially publishers (in truth, the whole thing likely comes down to the latter). That's a cynical view, more likely then not it just got away from him, and now this daft assumption that I somehow don't care because I was not responding to emails (in reality I had a killfile in place, never saw the emails- lack of response does not equate to consent though). Either way, read the emails.
So I went digging and finally found one of his active email addresses. Nice little discussion ensued, with me requesting not a redaction of the articles but to cease and desist usage of my name since it was my name, not random collision. Yanking the articles down Pete would never go for (apparently it is drawing readers), further the articles along with my name had already spread through the web. No single point to yank existing offenses, so it's a moot point. Real problem is preventing further use. Getting pist off and demanding something that won't happen serves no purpose, taking the long view was my approach- it was agreed that the name would be dropped from that point on. Bit surprising, but Pete's aparently decided afterwards to do otherwise, leading to this entry. The thread of emails are available here, should anyone doubt the truths above. Sorry for the mbox format, but it's the easiest way to bundle it out, and dump the relevant emails (quite frankly I don't care to spend any more time on this then I've been forced to).
Posing a rather interesting question for readers- how much is your name worth? Name the dollar amount that is enough for you to keep your mouth shut, and not stand up and scream "This is bullshit, that's my name!". Pete offered the somewhat princely sum of 45% of sales to not shake the boat. Pete has stated/claimed he has signed a deal with a European book publisher for sizable chunk of change (claimed million over the contract), so assuming he's not completely full of it, tempting. It's very fricking tempting, at first glance money for nothing until you realize that you're stuck with the name as Pete's son Gregory Douglas likely could comment on (not that it matters much living in Freeport, and working in a realm that isn't intricately tied to the web). Perhaps it's just an offer to share the wealth, the karma, although one must question why. Regardless, emails are available, you decide, this is merely my disclaimer.
Bit lopsided deal, with me as the loser from where I'm sitting. That and straight to the point, it's my damn name and reputation, I make the decision if I want to trash my name, not offered a consolation prize cause it's already occuring. I view the situation as a bit akin to if I ripped off someones car, wouldn't give it back, but was willing to give the person some cash for the vehicle. Problem here is it still is that person's name. Course the exception to this is you can't trademark a birthname, which is why you see a collection of porn stars with names of actual actors. The mainstream actors have no legal recourse to block it, unless the name they've popularized was a stage name, which can be trademarked, as can be pen names. Short version, there are legal recourses, but from what I've gathered the likely outcome of it would be a disclaimer from the publisher that I'm in fact not the actual author they're contracted with.
Ironic in a way that by opening my mouth, I'm binding the supposedly seperate Brian Harring's together, and getting sullied a bit via the questions of veracity Pete has already raised in using my name. That's not counting the annoyance of being forced to involve myself in cruft and potential loons (think people debating conspiracy theories without logic, strictly rhetoric). Whole thing is whacked but hey, you choose the company you keep, and there can be consequences of those you choose as friends (namely, they might slip a shiv in your back when you don't expect it).
This whole scenario does make me wonder what's it like for someone where a random pen name happens to match up to an actual person. Plus side, such a situation doesn't have overtones of loss of power (real or otherwise) to another you know. Legal games can be played, but frankly it's fighting from a point of weakness. Internet is vast, you stomp out abuse it's possible for the abuser to shift to a new location, and start anew. Should you doubt that, consider how spammers used to work before zombie nets proved themselves as a more efficient distribution system.
Pretty much from my viewpoint you're left continually on the defensive, which means you'll lose. The defender is stationary, the attacker is mobile, so you're pretty much boned. About the best you can do is stick out a disclaimer with a bit of ruckus, and move on. There's no point in spending time trying to police other's actions on the web, take your stand, and let chips lie where they may fall.
Final Summation: considering the fact that I've known Pete for several years, the unlikely chance of my name somehow winding up publishing news on his site, and the supplied emails, two things I can state my opion on regarding this insepid mess someone stupidly forced me to involve myself in-
Everything related (offending articles, and related email) I've jammed away in a directory should people suddenly decide not all publicity is good publicity. I won't be appending further articles since I have no intention of getting locked into stupid games of tracking what Pete is attempting further. It's just a waste of my time.
Frankly, must say I'm a bit pissed off regarding this whole thing. My general view of others actions is that as long as it's not affecting me (whether long term or short term), it's their choice. If someone want to spend their days stoned out of their gourd, I really don't care, although I may not agree with it. If you want to live as a luddite, kudos to you, just don't force your choices/consequences on me (I ain't giving up my computer dang it).
People can pretty much stick up whatever they want on a website- aside from legal concerns, readers decide if they believe it or not based upon what they want to believe, and occasionaly via critical thinking (crazy I know, but it happens sometimes). That said and done, my name turning up as an author on tbrnews is a bit much and forces me to step in and call bullshit. All in all, pretty damned idiotic, but neh, further fodder for people to argue without facts I spose (something y'all can kindly leave me the hell out of). In short, I didn't write the articles, see supplied emails for confirmation of this.
Update: Aparently there is another Brian Harring who is the real author- amazing the crap you have to go through to clear up a point- must admit the clarification from their end was pretty much a middle finger, but neh, so it goes.
Been pretty thwacked for time hunting for some more work, finishing current work, and general gentoo stuff. Keeps life interesting I spose, although I could definitely do with a few more hours in the day (probably would use 'em for sleep at this point).
Meanwhile, portage management is changing. Carpaski has stepped down as Operational Manager/Release monkey/Project Lead/Contributor, and Jason Stubbs (jstubbs), Marius Mauch (genone), and myself have taken up the reins. Pretty much continuing on with 'power' split between the three of us. Jason will be handling stable releases, I'll be managing the 2.1 alpha release when they hit (related note, a release policy is available). So... lots of happenings portage wise, although I've haven't had the time I'd like for any serious portage work as of late due to other projects. Fortunately that's a temporary distraction, still is annoying not having the time to finish off code I've got brewing locally. We also have a preliminary near future roadmap available.
Lance Albertson (Ramereth) hooked me up with access to some available gentoo hardware a
while
back, and have been abusing it towards distfile diffing (glep 25) ends, although
been
experimenting with sync'ing. Per hour delta avgs around 6.4kB in my testing of the last 11 days, which isn't too shabby. If you
synced from UTC 08 on 04/09, to now (04/20 08 UTC), it would require a download of 1.69mB, while *just* for rsync to
get to the point where it could start pulling down updates requires 2.55mB. For the sake of arguement, figure that the data pull
would be 1.5mB, if it were actually bzip2 compressed (it's not). So... like I said, not too shabby.
Rsync is damn fine for doing updates when you don't know what base version you're working from; stats above demonstrate that if you
can pin down what the base is, and pull pre-generated patches the bandwidth savings are pretty large.
Related to above commentary, said gentoo hardware (egret) is being abused to generate the daily snapshot deltas, detailed in an earlier blog posting. Not much new to report, still have a bit of work to do adding in uncompressed verification for the reconstructed snapshots (need to do that soon since bzip2 v1.03 is ~arch in the tree, and when it hits arch relying on checksumming the results of the local compressor becomes an issue). For those who were waiting for emerge-delta-webrsync to wind up in the tree, well, you have no further excuse now. I added emerge-delta-webrsync v1 to the tree on monday. Those too lazy to follow links, figure around 80-100kb avg per day download for syncing. Still have work to do on it in terms of cleansing distfiles after itself, and the aforementioned uncompressed md5 issue, but it's quite usable now
Hacking the hell out of mirror-dist as required for osprey.gentoo.org has cut into time pretty heavily as of late. Plus side, code is pretty much ready to go to phase out around 21gB out of 57gB for distfiles mirroring requirement. Frees up a lovely chunk of space (36% roughly), and will continue to keep the mirrored distfiles neat and trim once fully deployed. Kicks a bit of ass if I do say so myself, especially since it frees up space for patch mirroring. A lot of space, since if the patches available are carefully monitored and pruned, space required weighs in at around 3gb last time I went and generated all available diffs. That... seriously sucked. My box was basically punished for 3 days diffing around 40gB total. Quite happy I can abuse egret for it, since the hardware I'm using now has a lot less umpf, and 5-6 days of a process consuming all available IO doesn't sound too fun to me...
Meanwhile off to bed, after hitting the gas station for the 2 for $1 hot dog deal. Sad, but haven't hit the grocery store lately,
and doing it now at 4 am seems a bit whacked when I'm up in less then 6 hours :)
Hopefully I'll update the blog a bit more
frequently then I have over the last few weeks, 13 days is a bit
of a stretch...
Seems to have died down a bit, but kind of crazy to see fortune-mod-gentoo-dev (random insane quotes from #gentoo-dev) being used as blog comment spamming, for example here (comments section), here (15th comment in), and here. Swell. Joys of having a name/nick that isn't yet tightly bound together, resulting in lots of weird spam/crazy stuff getting matched by google I guess :)
Read about this a while back, and a question about why hard real time was important in #gentoo-embedded got me thinking of this again. Way back in '85, Atomic Energy Commission Limited and CGR got together to create a medical linear accelerator- just as it sounds, used for delivering radiation doses for chemotherapy. They created the therac 20, and therac 25- both had a software bug, except the later lacked hardware safeties, allowing the software bug to cause some seriously bad mayhem.
For example, the first case of the bug involed a patient getting in the range of 15,000 to 20,000 rads, when the avg rad dose was around 200. Bit of a difference there.
The report on it is really worth the read, and is available here. Figured I'd direct my readers (do I even have readers? :) to it.
Everybody hits software bugs- it's basically an accepted thing in the industry... the consequences of said bugs can be pretty horrible though.