Along with setting up NTP and SNMP on all the servers here, I've also been documenting the network here at The Company (not that Dan hasn't done that, but he's responsible for the overall network and tends to focus on the switches and routers—I've been concentrating on the hosts I manage plus the switches they're hooked into). In doing this, I came across a drawing program for Unix called Tgif, an Xlib based (woo hoo! no huge external libraries to worry about! Yea! No insanely large amount of dependencies! Woo hoo!) interactive 2-D drawing program.
I've spent the past few days with it, drawing the network—the rounded rectangles are network switches (the one in the upper left has two VLANs, one with 18 ports in green, and a second one with 6 ports in blue), the larger black squares are computers. Red squares designate off-page (as it were) connections. Red lines designate switch-to-switch connections. The purple square encloses the office machines and the turquoise squares are some comments about the two upper switches (yes, the image is shrunk).
But it was only when I was writing this entry did I realize that Tgif has hypertext capability (has since 1994! Ken! Why didn't you find this program?) and I've been going through the demo pages with the Tgif program since (and want to go back actually … so … um … end of entry).
Woah! I just realized that today is
Also, I've had several people just walk up to me and say, “Sync the mayo!” At first, I found it odd, but after the third time or so, I thought that maybe it was some sort of code, and I tried replying “I have a red pencil,” but that just made the person saying “sync the mayo” get a confused look on their face so I gave up on that phrase. I also tried replying “the sandals are very comfortable,” but again I got a glassy look.
I'm not even sure how to respond to “sync the mayo!”
I came across this rather forgettable article about Google's Web Accelerator. Normally, I would simply skip this and move on, but since Amanda (in a private post on her LiveJournal) asked about this, I figured I could answer her questions here (and entertain you people).
So anyway, here's a (nearly) point-by-point rebuttle to Think before you Worship:
There's a hardcore group of folks on the Internet who assume Google can do no wrong.
Yes, and they all work for Google.
To their credit, Google has spent years and millions upon millions of dollars cultivating this image, so I can understand where it comes from. The average Internet user seems to believe if they can't personally see a direct connection between ad revenue and a provided service, then the company responsible for such a product is some blameless, holy creature sent down by the good lord himself. “Well gee, Google offers all these services which don't cost me a thing, therefore they've got to be an unquestionable, infallible entity in my book!” the greasy nerd with 10-pound glasses exclaims while typing “MICROSOFT SUCKS” into his Google search bar. “Clearly this company only produces things for the good of all humanity! They aren't concerned with trivial things such as making money, dominating the marketplace, or boosting their stock prices!”
No, they are. If they weren't, they wouldn't be around for long. Any company that doesn't make money won't be around for long, and I'm sure that many companies would love to dominate the marketplace. Also, it's only public companies that want to boost their stock prices, and Google has only been a public company for about a year, if that.
This type of idiotic, blind fanboy mentality has led to where we are today: the Google Web Accelerator.
Funny, I thought that argument lead to us being in Iraq.
Millions of drooling bandwagoneers across the globe salivated at the thought of using yet another Google-produced product, once again passing over the fact that every business decision in Google's headquarters—and I repeat every business decision—is made with the bottom line of earning more money.
Since when is this news? Every business decision is made with the bottom line of earning more money—that's what a company does! It's what it exists for. Now, private companies have a bit more leeway in how they go about making money and what they can do with that money; public companies less so, since by law, they must incrase shareholder value, and that generally means anything to boost the stock price or dividends paid out.
True, bad decisions do come about (like the merger of HP and Compac) but even then it can still be traced back to the bottom line of earning more money, only in some cases, it's best to ask who's bottom is earning more money.
Google doesn't give a shit about you or me, unless you or me has the ability to give them money in some way. This is how every corporation is; they don't care about anything except making a buck.
Well, technically, a public company doesn't care about anything except for increasing shareholder value—they can't care about anything else by law! Private companies (like I said) have a bit more leeway since they aren't under public scrutiny (and that can be a Good Thing™ or a Bad Thing®).
Yet Google's history, their expensive team of lawyers, PR people, and viral advertising, has allowed them to ascend past the image of a generic corporate entity with the same bottom line as every generic corporate entity in the world. Such grace and good fortune has allowed them to release the Google Web Accelerator without facing a single shred of public scrutiny. So what does this incredible free new service accomplish?
Really? This isn't a publically viewable page? Wait a second … am I using the Google Web Accelerator? Oh wait … I'm getting ahead of myself here …
Google claims it magically speeds up your Internet connection, and for the average user, it does. This is done by caching entire websites on Google's servers, passing copies of any page a user visits and sending them to Google HQ.
Two points here—one, Google has already cached sites for their other major operation (search, remember? Also, anyone can view their cached version of a page by clicking on a link from their results page, and that fact could be used to survive a slashdotting) and two, they fetch the page on your behalf; you don't send them a copy at all. Does the author not understand basic web caching strategies? Or is he exagerating for effect?
Basically they're just creating carbon copies of everything you read, every site you visit, every image you load, and storing them on their servers, under the idea that you will have a faster connection to their servers than you would to whatever website you're visiting.
Well here's the problem, folks: everything you view is now owned by Google.
And if that is true, then you don't have to worry about this as the likes of CNN, the New York Times, Time, the RIAA, the MPAA, Playboy, Hustler and scores of other companies will nuke Google (from orbit—that's the only way to make sure) for even attempting to assert ownership of anything that is viewed by users of its service.
Do you read email? Well now Google reads your email, and now the entire world can read your email.
Yes I read email, but no, Google doesn't read my email—at least, it doesn't read any email not addressed to firstname.lastname@example.org (and frankly, while I have a GMail account, I don't bother using it at all).
But that's besides the point—even if I did use GMail, that doesn't mean that the entire world can now read my email (security incidents aside, and even then, everybody in the world would have to actively exploit any hypothetical security hole in order to read my email at Gmail).
I'd like to know how, say, Dad, could get my web viewing habits from the Google Web Accelerator (assuming I used said Google Web Accelerator, and if he were interested in reading what I read). Again, the author either doesn't know how the web works, or is simply trying to scare people from using the Google Web Accelerator.
I'm having to pay a coder just to figure out how to prevent Google from caching all the webpages on our forums.
There are a number of ways to prevent Google from caching your webpages,
and what method you use depends upon how much you trust Google. You could
try setting the
header. If Google uses a unique agent string (and Google already uses a
unique agent string when crawling a website for indexing) you
could check for that. The Google Web Accelerator may also come from a
set of known IP addresses and you could certainly block on that (from within
your webserver program, on the server itself or even at your site's main
router). If the coder you hired hasn't come up with this, then you're
wasting your money.
Why is this a problem? Well first of all, it's a giant security hole, as private forums for mods and admins can now be viewed by anybody. Thanks Google, thank you very much for sharing our sensitive information with the entire Internet, without even giving warning or notice to any parties involved!
Secondly, our forums offer a private messaging feature, where users can send messages to each other which can only be read by them. It's like AIM or ICQ, but through a webpage. If you're using Google's Web Accelerator—guess what?—now anybody can read your private messages! Cookies, logins, sensitive information, private messages—they're all stored on Google's servers now, and they're all available for anybody on the Internet to read.
I hate to break it to you, but if you think for a second that your
AIM or ICQ traffic is private then I have some nice
swamp land to sell you. AIM and ICQ go
through centralized servers and I'm sure that the technicians there get off
chat logs of the rich and famous. So again, nothing new there.
But again, I fail to see how (short of an exploit or some random Google engineer giving this out) this allows me to read your cookies, logins and other sensitive information.
So how can a company like Google get away with this?
Because the Scooby Gang have yet to unmask Sergei Brin?
If Microsoft had released something similar, you can damn well be sure millions of angry, greasy Lunix zealots would be carrying torches and pitchforks to Bill Gates' house. The reason Google can so brazenly trample over the entire Internet is because people let them. They spent so long carefully crafting their pristine Internet image and reputation that all their actions are now seen as faultless, for the good of all humanity. Google surely wouldn't do anything bad or evil! After all, their company's motto says so! You know, that company motto they paid a legal and creative team to come up with.
From what I understand, Sergei Brin and Larry Page wrote their company motto, and insisted that go into the SEC filings prior to their IPO (they also pissed off many brokers on Wall Street by making them do actual work—eeeeeeeeeeeevil company that Google).
For the company that makes millions upon millions of dollars. For a business fueled by capitalism, just like every other corporation in the history of the universe.
Making money isn't bad. The pursuit of money isn't bad. Money isn't bad. Failing to question companies that prey on you is bad. Google doesn't care about you or me or the Internet as a whole, unless one or all of these things can somehow make them another dollar. They are just like every other company and should be treated as such. Microsoft receives strict scrutiny for every move they make, which is a damn good thing because it keeps their company in check.
I wish, but I digress …
Google doesn't, and this is what led to the release of their Web Accelerator program. Bending over to worship at the feet of a company just gives them a better chance to step on your back and boost themselves higher. Google is just another Microsoft without the bad reputation.
I can see what Mr. Kyanka is worried about though—Google gaining access to content that it otherwise normally would not be able to see, and he's obviously worried about it be indexed into Google's search results and that is a cause for concern, but not for running about screaming “the sky is falling!” I think the engineers at Google are smart enough not to mix the content they get through their Web Accellerator with their search engine caches—so far, Google seems to “get” the web.
Although it's fine at sucking up dirt, finding its way around the room and returning to its charging station, its real achievement is in not only getting the children to clean up their toys first but also tiring them out before bed.
This works with three magic phrases:
- “Roomba's coming out tonight. Clean up your toys or Roomba will eat them!”
- “If you can clean them up fast you can stay up to watch Roomba!”
- “Here goes Roomba. Don't let him touch you!”
Yes, I've been rather lax in the updates here recently. Most of the time I don't have really anything to talk about, or the desire to talk about anything. And the rare case when I do have something I want to write about, well, it's at a time when I can't (not near an Internet capable computer or involved with some work) and when I do get the chance, I don't feel like updating.
I've written about this before (and yes, I stopped and spent a few minutes tracking down the relevant entry, making me loose some steam in the writing process—sigh). I guess that if it were even easier for me to write an entry, I would be more inclined to write more often, but I'm still waiting for that “Perfect HTML Editor” that will do The Right Thing™.
Then there's the whole “not having anything to write about” bit, but that's not entirely true. There's (and here I go, tracking down the relevant entries again—XXXX brick wall—not to mention actual work—yet more sighage; now, where was I? … oh yes) the fact that you do indeed require RARP to install Debian on a Cobalt RaQ2 (only Linux dropped support for RARP in the 2.3 timeframe and now you need a userland program—and it was fun getting that to compile); I finished the network diagram for The Company (and have been keeping it up to date) and now that I've finished the physical layer, I have to now work on the IP layer (which is a bit of a mess, although I'm not sure how much of that is due to BGP issues, and how much of it is legacy issues). And I haven't had to deal with a control panel issue in nearly a month (but even I am bored with detailing the control panel issues) (as a further note, making that sidebar took a good ten minutes, and yes, I have no idea what I was going to write about next).
So … well … um … there we are.
The general opinion of “Revenge of the Sith” seems to be that it marks a distinct improvement on the last two episodes, “The Phantom Menace” and “Attack of the Clones.” True, but only in the same way that dying from natural causes is preferable to crucifixion.
Via too many sites to mention, “Star Wars: Episode III.”
Ah, Star Wars.
And George Lucas bashing.
Just like in May 1999 when Smirk treated the entire office to Star Wars: The Phantom Menace, he is once again taking the entire office (although this time around it's a much smaller office) to see Star Wars: The Revenge of the Sith—the midnight showing tonight (well, technically, tomorrow morning).
I've written quite a bit on Star Wars in the past (and my friend Hoade and I have gone on for hours about the films and how Luke should have fallen to the Dark Side and Han (WHO SHOT FIRST! … ahem) should have been the next Jedi Knight; from cynic to believer in three films), but now … I don't think I can even muster the energy to write about it (and it's not because of the previous entry—but because I feel George Lucas blew the whole story arc). I know Smirk is really looking forward to “Revenge of the Sith,” and from the previews I've seen, it looks good.
But then again, previews are supposed to look good.
The Bad? Well, it wasn't made clear up front that General Grievous wasn't a robot (or not entirely a robot), so the first time you see him he's walking with a limp and coughing. My initial reaction—a coughing robot. Cute (blech). Also, R2-D2 got on my nerves, fighting and rocketing his way through the first half of the film (really—apparently by the time “Star Wars” happens, he's run out of rocket fuel and has to be manually lifted into Luke's X-wing fighter). The rather quick conversion of Anakin Skywalker from Jedi Knight to Sith Lord (although the moments leading up to it were wonderfully done). Natalie Portman—C-3PO had more presence in this film than she did. And I'm having a hard time rationalizing why Darth Sidious had Count Dooku kidnap Senator Palpatine (if you've seen the film, or don't mind spoilers, I do go into some more depth).
Several of the fight sequences head off into woosia territory (woosia is the term applied to martial arts film where the participants in a fight run up walls, jump between buildings, walk along tops of trees, leap tall buildings in a single bound, that type of silliness) and in some cases, it just seems way over the top, especially given the participants.
Darth Vader's first Frankensteinian steps.
And the midi-chlorian issue is raised yet again in a conversation between Senator Palpatine and Anakin.
The Good? Ewan McGregor and Hayden Christensen (a bit less whiny than in “Attack of the Clones”), especially their duel on the volcanic planet of Musatafar. The scenery and special effects were wonderful (with a few exceptions, like the giant lizard Obi-Wan rides in the second half of the film). Ian McDiarmid as Palpatine.
The Eh? Yoda. Sadly, Samual L. Jackson as Mace “Jules Winnfield” Windu (if only!). The fight between Yoda and Darth Sidious (starts out good, then goes woosia). The fight between Darth Sidious and Mace “Jules Winnfield” Windu (starts out woosia, then gets better). The chemistry between Anakin and Padmé. And Bail Organa's ship, which was the original ship being attack in “Star Wars” (they recreated the sets which just looked so anachronistic compared to the rest of the computer generated sets it was jarring).
Overall, I liked the film, and I agree with the placement of it in the Canonical List of Star Wars Ordering. And hopefully, this will put to rest “Star Wars.”
I can only hope.
And younglings? Younglings? George, they're called “children!”
A while ago the light in the short hallway between my room and the rest of the house blew out. When I went to change it, the bulb came out, but not the metal portion that actually screws into the receptacle—it was stuck. And not knowing which circuit breaker went to which circuit, we left it until we could determine what went where (not that we trusted the labeling on the circuit breaker box, but as any programmer will tell you, comments can be misleading).
Then last week a couple of the ceiling lights in my room blew out and when replacing one of them, the bulb came out, but not the metal portion. And again, not knowing which circuit breaker to switch, I put off changing the light bulb until I got a few hours of daylight to work with.
You see, in order to get the metal bases out of the sockets, it required the use of a potato—you jam the potato into the socket and use that to unscew the light bulb base. But before you resort to such tactics, it would be wise to shut off power to the lights—I'm not sure about you, but I'd rather not hold a frying potato in my hand.
So wlofie and I spent over an hour mapping out the circuits in Casa New Jersey. Results were quite amusing—one circuit has branches to my room, the living room and the kitchen; another circuit has another branch to my room and the living room; yet a third services the hall bathroom, the other bedroom and a living room light (now you begin to see why it took so long to map this out). There were four circuit breakers that didn't seem to be used, so I left them off and as we find stuff that doesn't work, we can turn them back on (in fact, later on, we found that one of the turned off circuit breakers serviced an outlet in the kitchen).
Once we had things mapped out, it was then time to change the light bulbs.
The potato thing didn't quite work out.
So I tried carrots.
The carrots worked much better.
It was during the shower this morning that I realized what one of the turned off circuit breakers was for.
On the plus side, this is Florida—the water never does get that cold.
For the past two days I've been fighting software. Yesterday was Bit Torrent, one of
those peer-to-peer file sharing programs. Wlofie asked me if I could burn a
copy of Slackware
10. A cursory glace at the website showed that the only way to get
CD-ROM images (so-called “isos” due to their file extention
.iso) was through this thing called Bit Torrent.
I think I made the mistake of using the clients listed, and a second mistake by using the GUI client provided. I'm pretty minimal when it comes to GUIs, and I absolutely refuse to use the Gnome or KDE desktops (they tend to make a 1GHz machine feel sluggish).
I was much displeased when, by mistake, a stray click on the graphical Bit Torrent client launched this … monstrosity … of a desktop that took over the system (more or less). My background changed, thousands of extraneous programs started up, the beeping, the buzzing, the sluggishness of a Windows system … just horrible.
I ended up pulling the power cable in self defense.
Well, that was yesterday.
Today … today was worse.
There was the fight with Twiki (which we use as a company wide knowledge base)—who would have thought updating some graphics would be so XXXXXXX difficult? (all I wanted was to update the network maps, but … no … must stop before my head explodes on that problem)
Then Firefox. Granted, it
was version 1.0 so most of the annoyances are fixed in a later
version. Getting that later version? Leads us to the next piece
of software I was fighting—
yum—the RedHat package manager (as if RPMs weren't bad enough). I could
yum to work on my system here at The Company because
it complained about PGP keys
not being configured or something like that. It took P about ten minutes to
untable that mess, then it was off to install all 137,000 megabytes
of webbrowser and email client.
Meanwhile, I've been fighting to get this webcam installed in the datacenter. It's one of those stand-alone units that plugs right into the network, with a controlpanel (cough cough) and it can be configured to upload images to another computer either by FTP or SMTP.
It's been a huge mess—networking issues,
apt-get (Debian's package manager)
problems (to say that the MIPS based RaQs are a niche market is a bit of an
problems (the webcam apparently supports FTP in name only), cabling issues (the network cable I
crimped for the webcam was marginal and I had to recrimp it) and DNS issues (which involved some other
Cobalt RaQs and the Cobalt RaQ's wonderful controlpanels (cough cough) and parsing of e-mail (since
the webcam's FTP is
obviously borked, have to use SMTP to send the images—lovely—but that, unlike
But (thankfully) is now working, and managed to catch this joker trying to sneak off with some computer equipment:
So that's one thing off the to-do list.
And I'm finally calming down.
The one thing that did go right today? Burning CD-ROMs. As long as I finally got the Slackware
.iso images, I might as well tempt fate and burn some CDs. Oddly enough, with an external DVD burner hooked up via USB to my Linux workstation, I
managed to make a bootable Slackware CD with one command:
cdrecord dev=0,0,0 speed=4 slackware-10.1-install-d1.iso
(okay, there were three more CDs to burn, and yes, it wasn't the fastest of operations, but I wasn't looking for fast, I was looking for it works! And yes, it works)
So I am happy that worked.
Hopefully, tomorrow will be a bit better …
It's thundering outside, and I see the occasional flash of lightning off in the distance. There's an electrician installing some circuits in the data center (I'm watching on the webcam).
I've got a bad feeling about this …
Spring is homeschooling the Kids, and as such, she's been buying (or being given) used textbooks (of various grade levels) for use in teaching. She just now brought home a metric butt-load (a technical term for “a lot”) of books. And it's not like we can afford to be all that picky.
As Dave Barry says, I am not making this up.
Turning to the “Dogs: Devoted Carnivores” section, I read:
The dogs on the Ark.
How many kinds of dogs can you name in one minute? All dogs, domestic or wild, are called canines by scientists. Did you know that all dogs—whether domestic dogs like poodles, cocker spaniels, Irish setters, collies, and Great Danes, or wild dogs, such as wolves, coyotes, jackals, and dingoes [dĭng″ gōz]—may have descened from just one pair of dogs that Noah took with him on the Ark? At Creation God gave dogs the ability to bear young possessing a wide variety of characteristics. Through the years, the different types of wild and domestic dogs that we see today have developed. No dog has ever turned into any other kind of animal, however. Dogs always produce dogs, never cats, mice, or monkeys.
This book is heavy with a literal interpretation of the Bible. For instance, in a section on “Men Who Saw Dinosaurs:”
Perhaps the most interesting story in the Bible about a man who saw a dinosaur is found in the book of Job. In chapters 40 and 41, God Himself describes two magnificent creatures that Job knew about and had probably seen. The first creature is called behemoth [bĭ ‧ hē′ məth]. In Job 40:14-21, we read that behemoth eats grass like an ox, moves his large tail like a giant cedar tree, has bones as strong as iron bars, can drink up a river, and “is the chief of the ways of God,” probably the largest animal on land. The word behemoth means a “gigantic beast,” possibly like the huge Brachiosaurus [brā″ kĭ ‧ ə ‧ sô′ rəs] or Apatosaurus [ăp″ ə ‧ tə ‧ sô′ rəs].
The second creature God describes in the book of Job is leviathan [lə ‧ vī′ ə thən]. Chapter 41 says that leviathan lives in the water, has terrible teeth and scales, and is very strong and fearless. Everyone that sees him is afraid. Verses 20 and 21 say that smoke comes out of his mouth! Leviathan sounds a lot like the stories you've heard about fire-breathing dragons, doesn't he? Maybe some of those “dragon” stories are really true. Scientists have found certain dinosaur skulls that have a strange “bump” on the top of the head. These bumps or crests were hollow and were joined to the dinosaur's nose by tubes. Some scientists believe the bumps could have been used to mix together gases which burst into flames when they were beathed out into the air by the dinosaur. Leviathan must have been some sort of large, powerful, fire-breathing “sea monster,” maybe like the huge Kronosaurus [krō″ nə ‧ sô′ rəs].
And here I thought those “bumps” on the top of the “dinosaurs” were “breathing tubes” since these “creatures” lived under “water.”
And oddly enough, they don't mention by name any of the scientists that believe in these “fire breathing dinosaurs.” Odd how that works.
Then there's the other book she handed me, Matter & Motion in God's Universe (also a Beka Book). This one has an entire chapter on “Science vs. Evolution” and has a section on (and I swear I wish I was making this up) “The Fossil Record Supports Creation, Not Evolution.”
As if Kansas wasn't bad enough.
Remove the pro-Christian messages (and really, these are books on science, not religion or philosophy) and they're are all that bad. I do like how they provide pronunciation guides for possibly tough words and they do cover a range of topics. But it's hard to get past stuff like:
2. How might the Flood have caused the extinction of the dinosaurs?
I mean, the answer is—it shouldn't have! As God said to Noah, “of every living thing of all flesh” (by twos, or by sevens, depending). So the dinosaurs shouldn't have perished in the Flood, unless you want to consider that the Bible is, in fact, infallible and God really didn't mean “every living thing of all flesh” but I'm not going there.
Satire just can't compete in today's world.
Update on Monday, May 30th, 2005
Update on Thursday, June 2nd, 2005
Oh, and in case you're baffled because the robot dinosaurs are going to laugh about pinhead scientists and their extinction theories, then talk about how they got on the ark with Noah, two by two, then say they died because of the great flood, just so you don't have to look it up: the global climate changed drastically after the floodwaters subsided to one that specifically targeted dinosaurs for death. That's right: God made Noah and his family carry umpteen dinosaur couples on the ark for a year and feed them bales of hay and whole cows and shovel their massive turds, and then he killed them all right after they stepped off the boat.
Gotta love that God!
Update on Thursday, June 2nd, 2005
It's been awhile since I last wrote about this program, and in the time since it's seen a bit of work (it hasn't been a real high priority as you can tell). The color scheme has changed a bit since I last wrote, with green denoting white's influence over the board, and a red border now showing conflicting influence; the squares that are both blue and green show what percentage of influence each side has. I also highlight the movement of the piece that's under the mouse cursor (here the mouse was over the white bishop in the penultimate row) although sometimes it's somewhat hard to tell (without using totally garish colors).
On the chessboard shown, the black king is in check (easily seen since it's on a contested square) by the white bishop, which itself is in a contested square. The two gray squares in the top middle of the board are not under the influence of either side, even though the two white pawns can easily move into those squares, which leads me to the next thing I've discovered as I've been working on this: the pawns are the hardest to deal with.
Pawns move straight in one direction, but capture on the diagonal (except for en passant which I don't even want to deal with at this point) so calculating the influence of a pawn over the board isn't quite that easy, but I think I got it down.
And yes, I do need to rotate the board clockwise to match the orientation used in chess documentation.