Home
Stephen Harris' Friends
 
 [Most Recent Entries] [Calendar View] [Friends View]

Below are the most recent 25 friends' journal entries.

    [ << Previous 25 ]
    Tuesday, July 14th, 2009
    blarglefiend
         		8:39p
    And meanwhile...
     Today was also the first day back in the office for over a week. I remember why I hate it so much. Spent a lot of time trying to get my head around Munin and Zabbix by reading documentation and doing web searches to try and answer some questions of the "is it worth my while even installing this thing?" nature while my boss spent half the afternoon with a loud speakerphone conversation running.

    A reason to hate the current monitoring system: in the week I've been out of the office 10,000 emails came from WUG. I can not believe there were 10,000 genuine problems over that period. Having cleaned up those folders there were approximately 50 alerts that tripped my "UNIX" filter just in the afternoon. They were all of course of the "oh noes, something took too long to respond to a ping!" variety. I can not understand why my colleagues do not see this as being a significant (and fatal!) flaw.

    To put it into simple terms: a monitoring system which produces a huge number of false alerts is a monitoring system that is going to be ignored. Even if you pay people to sit there and watch the thing they will miss real problems.

    Am complying with the "no external email" rule by using my iPhone for that instead. Not that I get a lot of mail that isn't either spam or caught in one of my various filtering rules -- the only mail that hits my inbox these days is things from Real People, the odd bit of spam, the Crikey daily mail, and the odd commercial-but-not-spam message that I haven't yet filtered into the appropriate folder.

    Installed Zabbix. Goodness, what an odd bundle of software. Very... idiosyncratic. The documentation looks comprehensive until you actually try to install it, whereupon one realises that there are huge gaps. Shall poke at it further tomorrow, but am not holding out much hope.

    Will probably do a test Nagios instance -- not touching the one monitoring the Sybase servers! -- and poke at integration with Munin. That ought to resolve most concerns about scaling plus provide genuinely useful performance statistics. Remains to be seen just how much data the Solaris plugin provides as the demo site is presumably all-Linux. But it ought to be quite straightforward to munge the Nagios check_sybase plugin to get some basic stats for Munin, so that'll be nice.

    The other advantage of doing it this way would be that most Nagios host definitions become much much simpler. No more having to define every filesystem or database segment.

    Am also thinking a bit about how to integrate Splunk with Nagios. It seems reasonably obvious, given the "trigger shell script" option for saved searches. Instead of sending a cryptic email directly it can send a message via NSCA and let Nagios handle the notifications.

    It remains to be seen if I will be allowed to admit that any of this exists, presuming it all works as advertised...
    blarglefiend
         		8:25p
    Day One
     One day does not a substantive weight-loss effort make, but meh.

    The airline food arrived during the day. Bin night, so I went through the freezer and threw out all the "tempting" crap, replacing it with the airline food. So now the minimum-effort option is to eat that.

    Am back to using Calorie King. Shall see how long that lasts. But if it's to be believed then I can see why I'm still rather hungry -- less than 1300kcal so far today.

    Put in a grocery order for delivery tomorrow night, none of the usual "food" suspects. A couple of apples, a couple of bananas, some gear to try and repair some shoes, other randoms. And nabbed some coconut-covered-peanuts and wasabi peas to try.
    dilbertdaily 12:00a
    rlwinm 12:52a
    The truth is also a three-edged sword.

    http://rlwinm.blogspot.com/2009/07/truth-is-three-edged-sword.html

    But these are a different three from the canonical ones for understanding. Of course I'm talking about the RAIDframe stuff, because that's all I seem to get around to posting here. Anyway, there's what we believed at boot, what we know now, and what we want to be believed on the next boot. For the code I've written, they're fields of the same struct. For the existing raid(4) code, the information can be a bit more… scattered. (Making things slightly more fun: all the metadata for a RAID set is replicated on each component, so there's the question of what to do if there are non-fatal differences.) My SoC mentor has noted that things could use some reorganizing there, and part of me would like that too, but a much larger part of me says It's Working Code, Leave It Alone.

    This notion applies to fault-tolerant persistent data systems in general, really; there, as with RAIDframe, the first item is relevant only until some kind of roll-forward is done to clean up after a failure. In RAIDframe's case, this is raidctl -P, and it's a little more prominent because it runs in parallel with, to borrow a term, the mutator; contrast with wapbl(4)'s roll-forward, which is done automatically on mount, appears to be quite quick, and I assume blocks use of the filesystem until it's done.

    (In the other half of my life I'm looking at the literature on persistence, and it's almost odd how these two things are converging here.)
    Monday, July 13th, 2009
    syringavulgaris
         		10:28p
    Wunce Agane Meat Fale Me
     I returned this evening from the podiatrist, who judges that I have plantar fasciitis. Short form if you are suffering TL;DR: the connective stuff on the bottom of my (left) foot? It's inflamed and sad. This happens if you run, if you're overweight, if you're female, if you lug heavy stuff around, and if you wear high-heels or flip-flops (any shoes without arch support, basically). OH GOSH WELL I DON'T DO ANY OF THAT

    So. I am to take ibuprofen for a week, and I am to ice the sad bit 2 or 3 times per day. I am to do stretching exercises[1]. I am not to run. I am to try and keep off it, and not do things that bother it. And I am to wear sensible shoes with arch support. IN SUMMER. kill stab hate kill

    I am feeling whiny and oppressed; and also, if packing extra poundage contributes, how am I gonna lose it if I have to keep off my feet?! (The college pool is open, but I find laps so unspeakably tedious I can't stick to them... On [info]sweh's advice I am looking into waterproof MP3 players, which would tidily solve my problem, but they seem to be much expense for low functionality.)

    [1] One of the primary causes is tightness in the Achilles tendon and hamstrings. Yoga for almost a year? And I still have tightness there? what does it take
    janetmiles
         		9:47p
    dvandom
         		6:57p
    RotF Scalpel options
     Turns out there's two mutually exclusive ways to assemble the abdomen (well, thorax) of Scalpel's robot-arthropod mode. Two sets of pegs that can't be used at the same time.
    ( Ach, mein digits! )
    eaglespath 3:43p
    check-dist 1.3

    http://www.eyrie.org/~eagle/journal/2009-07/007.html

    I've been converting all of my software packages to Automake, which by and large is a significant improvement over hand-rolling makefiles. However, Automake does have the requirement that all files to be included in the distribution be listed explicitly in Makefile.am, and I've found it far too easy to miss bits of documentation or parts of the test suite that aren't always run.

    I therefore wrote this script to do a sanity check. It's a simple script that takes a tarball and a source tree and checks that the tarball contains everything that's in the source tree, excluding some files to ignore that are configured at the top of the script.

    You can get the latest version from my scripts page. I probably won't announce new releases unless they're significant.

    imc
         		9:42p
    Humour meme
     No idea how they calculated this, but:

    ( Haven't done one for ages… )

    janetmiles
         		3:49p
    charlies_diary 6:38p
    Back to the Moon?

    http://www.antipope.org/charlie/blog-static/2009/07/bad_blogger_no_entry.html

    I was thinking I ought to be looking for something creative to say here — blogging gets old, after the first eight years — when a bright young thing at $PUBLISHING_COMPANY emailed me to say: "on the 20th, it's going to be the 40th anniversary of moon landing day, wanna blog about where you were and what it means to you on our corporate soapbox?"

    To which I said "sure, but I was only four years old at the time ..."

    I hope to live long enough to be four years old that way all over again, some day.

    Why didn't Apollo stick?

    Well, for one thing, it was a stunt.

    Wernher von Braun and his colleagues didn't see it as a stunt, of course. They saw it as a stepping-stone, a valuable intermediate step in establishing a metaphorical beachhead in space. And rightly so (given the state of knowledge of how biological organisms handle zero gravity and high radiation environments at that time).

    Unfortunately, the real goal of the rocketry pioneers called for something a bit bigger than an ICBM. Funding their requirements ... not easy. The price of a launch to orbit scales roughly in proportion to the cube of the payload. A modern, miniaturized H-bomb weighs about 200-400Kg (more for the really big stuff, but they're not terribly effective militarily). An ICBM that can lob 3 warheads around the world is actually only lifting about one ton of payload, and it doesn't even reach orbital velocity: it's around the bottom end of a viable space launcher. Von Braun and his Soviet counterparts like Sergein Korolyov knew this well. Korolyov had the relative luxury of the USSR's plutonium shortage and lack of proficiency in guidance technology to help him justify building big rockets: these factors forced the Soviet ballistic missile forces to use single multi-megaton H-bombs (weighing several tons), which in turn needed a 300 ton monster like the R-7 Semyorka to achieve their military goals. The R-7 was big enough to throw a small satellite into orbit without upgrades; with incremental improvements, it has continued to evolve (long after its military obsolescence) into the R-7 family of space launchers, one of the safest and most reliable multi-stage liquid fuel rockets ever.

    But the United States had much better guidance technology, and more plutonium (and could therefore use numerous small, efficient, accurately-targeted warheads to do the job for which the Soviets had to rely on a few cumbersome city-busters). While the early US ICBMs were reconfigurable as space launchers, the really big job of going to the moon required something new — the Saturn series of boosters. And the Saturn V configurations that eventually flew to the moon and launched Skylab were far from the largest variants planned.

    Let me put it this way: the goal of going to the moon by 1970 forced design compromises on NASA. The objective of putting a man on the moon and bringing him back would be achieved using the lightest lunar lander conceivable — one so flimsily built that, on one occasion, an engineer working in the cabin of one of the LEMs dropped a screwdriver point-down and it punched right through the spacecraft's skin. The LEM weighed under 15 tons, fully fuelled; the ascent stage, in which the crew were to live until they returned to dock with the command module, weighed barely 2000 kilograms (plus crew and 2350Kg of fuel).

    Not only does the cost of putting a payload into orbit increase with the cube of the payload weight — this rule holds true in the opposite direction, too. Stick a LEM on the moon and bring the contents back? Easy. Increase the mass that the LEM brings back? Very expensive — the price goes up as the sixth power of the weight you're returning from the lunar surface (because you have to loft the heavier LEM into Earth orbit to begin with).

    Think about it. The real mission wasn't to go to the moon; it was to bring two astronauts and 100Kg of moon rocks back from the lunar surface and into lunar orbit (to rendezvous with the CSM stack for the journey home) — and it took a 3000 ton behemoth to accomplish this. Launching a bigger, more useful LEM (one that could carry 3 or 4 astronauts to the lunar surface, along with a decent-sized rover and supplies for a couple of weeks) would have added tonnes to the LEM payload ... and hundreds, if not thousands of tons to the launch stack. With cost scaling as the cube of the vehicle mass, you don't need to be an accountant to realize that the US government, stuck fighting a war in South East Asia, wasn't going to give NASA the money to build in even one kilogram more of payload than was strictly necessary. Indeed, the mushrooming weight of the LEM (it gained about 15-20% during development) threatened to jeopardize the whole mission profile — except that the Saturn V performance exceeded expectations. The per-launch cost of even a minimal Apollo moon shot was $431M, in 1967 dollars — call that $5-10Bn today.

    So: the original Apollo moon shots couldn't have easily scaled up to achieve more — not without throwing a lot more money at the program in 1968-70, at a point when NASA was already consuming 0.5% of US GDP. For comparison, NASA's budget in FY 2008 was $17.3Bn; if NASA was funded at 1967 levels today, its budget would be closer to $75Bn.

    Is NASA capable of going back to the moon?

    I want to believe. But ...

    First the argument in favour. The Soviet manned lunar program (cancelled in 1969) were running on less than a tenth of NASA's budget; they nevertheless got four flights into the test program for the N1 before it was cancelled, and may have been on the edge of solving its engine problems. Today, with better simulation and modelling techniques, new materials science, and much better electronics — not to mention 30 years more experience in space exploration — we ought to be able to design and build better (and importantly, cheaper) spacecraft.

    Moreover, on the basis of the current estimates, a lunar round trip in Orion/Altair will represent a huge bargain compared to Apollo — delivering 60% of the lunar astronaut surface duration of the entire Apollo program, for a tenth to a fifth the price of a single 1960s moon shot. Even if it undergoes a 100% cost overrun, it'd be cheap compared to Apollo.

    But there are problems. Today we lack a vital resource that both Wernher von Braun and Sergei Korolev took for granted: thousands of engineers with the experience of designing, building, and launching new types of rocket in a matter of years or even months. We used to have them, but some time in the past 40 years they all retired. We've got the institutions and the data and the better technology, but we don't have the experience those early pioneers had. And I'm betting that the process of rebuilding all that institutional competence is going to run over budget. While NASA's Constellation program might work, and while it could deliver far more valuable lunar science than Apollo ever did, it will inevitably cost much more than NASA's official estimates suggest, because it's too big a project for today's NASA — NASA, and indeed the entire space industrial sector in the USA, would have to grow, structurally, to make it work.

    Constellation won't survive a 100% cost overrun — or even a 25% one. Instead, it will almost certainly be cancelled. The fiscal realities of the second decade of the 21st century are horrendously worse than those of the sixth decade of the 20th. The sixties were a boom decade; a better comparison would be the stagflation era of the seventies (never mind the Great Depression). The Shuttle program survived the 1970s but only just — its budget scraped through by a single senate vote at one point. And the Shuttle's specifications were mutilated by the political need for it to support an Air Force mission (one that it never flew, but which nevertheless ultimately doomed the Columbia). Project Constellation has no such USAF mission to cling on to for dear life when the budget axe is swinging.

    And so I very much fear that I'm going to have to stay fit and aim to live for a very long time if I want to feel four years old all over again.

    janetmiles
         		1:39p
    dilbertdaily 12:00a
    blarglefiend
         		1:46p
    Vague attempts at productivity
     The weekend past saw me scouring through bags and boxes and piles of stuff seeking out remaining items that my ex left behind when she moved out a year ago.  I believe it is now all collected into a couple of plastic bags to add to the things I already knew were waiting, stashed under the stairs.  Pick-up is being arranged.

    It also saw me searching for tax paperwork so I can finally get that great big mess sorted out.  I am torn between trying to handle it myself -- the actual tax returns are fairly straightforward for the most part -- and paying an accountant to take it off my hands.  The complicating quirk is that my having been honest and declared a small bit of work on the side way back in 1999 has left me with the tax office insisting that I am running a business and assessing a few thousand dollars per year on that basis in addition to the taxes already paid, so they erroneously believe that I owe them another $20k.  I have previously tried to resolve this by returning their big batch of forms, all filled in to indicate that for each period in question there was no business, but they simply returned new assessment notices that ignored the "new" information.

    So an accountant or tax agent might be the way to go, if only because they should know how the hell to handle this.  Trying to talk to the tax office has been fruitless, "we are not tax advisors", etc.  I do at least have some recommendations in this direction.  Have most of the necessary paperwork now, just need to get reprints from Monash and the current employer for last year.

    Incidentally, Monash is the only place I've ever worked where they demand payment in return for this.  If by post, then cheque or postal order.  This has also been a stumbling point in the past as I do not own a cheque-book -- the cost of doing so is additional taxes on one's bank account -- and almost never enter a bank branch of post office.  Shall combine this task with passport renewal once I find a suitable place to get a new passport photo.

    The sorting through the contents of the wardrobe in search of Leah's leftovers made me face the amount of pretty decent clothing I have that I simply can't wear due to weight gain.  It's all stuff I was wearing before this whole MS thing kicked off.  I can either lose some damned weight and be able to wear the nice things I already own, or I can buy more nice things.  The less-easy road is in this case the more constructive, so I have renewed my enthusiasm for airline food and put in an order for delivery tomorrow.  In truth the food isn't bad, and it's not like my present diet is particularly brilliant or varied.  While I would be eating better food if I were to cook it myself, the reality is that I'm not very good at that and find the whole business daunting enough for that to be a blocker.

    So, airline food it is.

    I should like to be back around where I was six years ago by the end of the year.  That'd be about 20kg lost, which is perhaps a little optimistic but not completely implausible.

    Today I am taking the day off, having not got to sleep until 4AM and then having a fairly broken slumber.  There was, at least, a really nice dream in which I was being kissed by a strange -- in the "never seen her before" sense rather than that which might more usually apply to women who kiss me -- woman who then proposed marriage.  Odd and probably liable to cause considerable consternation if it were to happen in reality, but perfectly pleasant in the context.

    Also, trying a new prose style.

    Sunday, July 12th, 2009
    syringavulgaris
         		10:07p
    In which there is Creativity
     So, occasion: [info]nedlnthred's boyfriend's birthday; a gala affair where the guests were requested to come in black tie, lingerie, or togas. Being us, we chose Door Number Two; something Moulin Rouge-inspired, perhaps with a dash of the ahistorical but really nifty courtesan outfits in Dangerous Beauty.

    It so hap'd I had need to shop for corsets anyways, so I went down to Purple Passion, and they had something in my size that looked good on me, so I snarfed it. (It's claret silk with black/burgundy rose tracery.) I can't afford it, the more so upon realizing just how little we're actually getting out of the house sale (math is hard! let's go shopping!), but eating ramen is cheap and slimming, neh? But then, what to wear for a skirt? Thus I spent Saturday at Beth's, while she was making a corset from scratch, oh yes she did, figuring out how to turn my bias-flared black skirt into an Object of Greater Interest.

    I took some of the scarlet silk taffeta Beth had bought for her skirt, cut it into seven 1.5" wide strips, and tacked them vertically on the skirt. Then I took some black lace (she needed the border motif; I used the center bit) and made a lace overskirt, which I swagged up along the front to match the bias flare, attaching it on each side with pink satin ribbon roses that I dug out of a corner. So that, and the corset over it, and garter belt with fishnets, and knee-high black satin boots with stiletto heels, and delicate pink-rose motif pendant necklace with matching earrings; and I did my hair in a coronet of braids with pink rosebuds from [info]sweh's garden woven in. I am pleased to say that, for once, the end product of my labors came out pretty much as I had visualized them. It was a difficult birth, but the delivery was even mostly on time.

    Lessons learnt:
    1) Wow, a dress mannequin really does make it easer to faff around with what you're doing.
    1a) But it's hard to faff when you don't want to be cutting your tolerably large bolt of fabric yet, since you haven't worked out how large a piece (or pieces) you will need for the task. This seems to require a level of visualization higher than my current tech level.
    2) My machine-stitching skills are still kinda crap (though, in my defense, Beth's machine is not unlike a very fussy Thoroughbred).
    3) Taffeta catches the light in markedly differing fashion depending on which direction you cut it in. (This caused me to waste some fabric {though it can probably be used for bindings} and about 40 minutes.)
    4) I want to play more with shiny pretty things, which regrettably my usual historical efforts do not allow for.
    5) I can drive an hour+ in a corset, because I am awesome.

    Current Music: AC/DC, "Shake A Leg"
    janetmiles
         		9:17p
    ewx
         		6:55p
    VCS 0.6

    I have released VCS 0.6.

    What's VCS?

    Do you find yourself typing bzr commit in Git branches, or cvs diff in your Perforce workspace? Then perhaps VCS is for you.

    VCS is a wrapper for version control systems. It presents an essentially uniform interface to the user, allowing ‘muscle memory’ to use vcs commands rather than adapt to the version control system currently in use. Of course, the downside is that you have yet another three-letter command to start using!

    The supported systems are Bazaar, Git, CVS, Subversion, Mercurial, Darcs and Perforce. More could be added. The supported commands are add, annotate (blame), clone, commit, diff, edit, log, remove, revert, status, update. Again, more could be added.

    Where Do I Get It?

    http://www.greenend.org.uk/rjk/2009/vcs.html for source code, .deb files and Bazaar branches.

    What's New In VCS 0.6?

    ( Various things )
    jwz
         		10:14a
    RFID passport wardriving

    Obviously the workaround is to stay away from Fisherman's Wharf.

    Zipping past Fisherman's Wharf, Chris Paget's scanner downloaded to his laptop the unique serial numbers of two pedestrians' electronic U.S. passport cards embedded with radio frequency identification, or RFID, tags. Within an hour, he'd "skimmed" four more of the new, microchipped PASS cards from a distance of 20 feet.

    "There's a reason you don't wear your Social Security number across your T-shirt," Albrecht says, "and beaming out your new, national RFID number in a 30-foot radius would be far worse."

    But Gigi Zenk, a spokeswoman for the Washington state Department of Licensing, says Americans "aren't that concerned about the RFID" in a time when "tracking an individual is much easier through a cell phone."



    Current Music: Emergency Broadcast Network -- Station Identification
    vatine
         		11:01a
    #48, "Blood of elves", Andrzej Sapkowski
     Previously unread.

    This is, if I my memory serves me right, the second Polish F&SF author I've read (the first being the unimitable Stanislaw Lem). But, with that out of the way, let's plow straight into my perceptions of the book.

    On the whole, quite readable. There are a few things that weren't necessarily top-notch. There's the occasional skipping-in-time tha isn't clearly marked (it's obvious what's happening, but there's no real forewarning). Not ideal. But, I can see why it was done.

    The second thing that bugs me about this book (and that may well just be in the English translation, as I can't even order beer in Polish, I dare not attempt to read a book in it) is that it just stops. Suddenly, there are no more pages, but there's no obvious reason why it ends where it ends. It gives the impression to be the first in a sequence (looking further, it does indeed seem to be the first, chronologically, in a series), but ether's no obvious indication on the book that this is the case.

    Readable? Yeah, on the balance, it is. Do I regret buying it? No, but it was on a 3-for-2 deal and I may just about have been miffed if I'd paid full price for it. Buy more in the same world? Maybe. I haven't ruled it out as an impossibility, but I can't say I'm itching to buy another one.
    lnr
         		10:50a
    The weekend

    Nice quite Saturday with much geeking, and finishing the book I started on Friday, with curry and The Commitments in the evening. Slightly spoiled by waking up on Sunday and discovering computer won't boot. We unplugged everything, opened the case, checked nothing had come loose and then tried it again and now it's booting OK. Stupid thing.

    We intend to spend the afternoon at Shelford Feast.

    dilbertdaily 12:00a
    ptomblin_lj
         		2:38a
    5th Annual Armond Bassett Canoe/Kayak Race

    “It could be worse – it could be raining”
    – Marty Feldman, Young Frankenstein

    One of my top goals this year was to finish a 10 mile kayak race, and today I did it. The Armond Bassett race takes place here in town at the Genesee Waterways Center, a place I’ve paddled a lot in the last year. Maybe I’m not being fair to the organizers, but in some ways the AB seems like a canoe race that reluctantly allows kayaks to participate. I can’t really explain why I feel that way, but I do. Maybe it’s because on the second day (tomorrow), they have canoe races that *don’t* allow kayaks to participate.

    Read the rest of this entry »

    Originally posted at Rants and Revelations
    Saturday, July 11th, 2009
    dvandom
         		7:38p
    Play CoH? Got an Odd Lots nearby?
     Odd Lots has gotten in a bunch of old City of Villains box sets. The codes in them are STILL VALID. So it's $6 for a month, plus you get a CoH Heroclix and the various CoV paperwork (Rogue Isles map, instruction manual if you want one). I only bought one because I wasn't sure the code would work and figured the other stuff was worth $6 for me to get one copy, but if there's any left next time I'm in Topeka I'm cleaning them out.
    janetmiles
         		5:01p
    linodestatus 7:17p
    [ << Previous 25 ]
A Slave Named Stephen Harris   About LiveJournal.com

Advertisement