TTG Home » Robert Home » Daynotes Journal Home » Journal for Week of 7 January 2002
 |
Week of 7 January 2002 Latest Update: Friday, 10 May 2002 12:57 |
|
| Linux Chronicles | |||
|
Monday, 7 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 10:45 - The
start of another week, and my to-do list runneth over. I have PC
Hardware in a Nutshell chapters in progress, a lecture for the
astronomy club that I haven't finished writing, Linux to install on one
system, Windows XP to install on another, and a system or two that I need
to build. I also need to spend some quality time on Barbara's and my
project to do a series of resource books for fiction writers. Well, you
get the idea. More stuff to do than time to do it.
And yet another intriguing possibility beckoned the other day. Pournelle called, and at one point during the conversation he mentioned that he needed to put some time in writing fiction. He has the Burning ... series with Niven, as well as another book due (or overdue) in his Janissaries series. Then there's his Falkenberg's Legion paperback series. I told Jerry that if he didn't write another book in that series pretty soon, I'd write it myself. Jerry said, "Okay, why don't we co-author it?" Arrrrghhh. My bluff was called. I told Jerry I'd love to, but I simply don't have the time. Maybe I should make the time, but the real problem is that I don't know if I'm a story-teller.
I knew that 14mm Pentax XL I bought for Barbara was a large eyepiece, but I finally did a side-by-side comparison with some other eyepieces. The one on the left is a standard Orion 25mm Plössl. At center is an Orion 30mm Ultrascopic (identical to the 30mm Celestron Ultima). That's a reasonably large eyepiece itself, with an eye lens that's as big as the objective of a small binocular. The 14mm Pentax XL on the right is in its own class. I can see why some people have to rebalance their scopes to use Pentax eyepieces.
These things are large enough that they won't fit conveniently in our eyepiece case. I'm going to have to do some thinking about this, particularly if we end up with several of them, as I suspect we will.
The freezing rain did arrive, although the temperature was high enough that it soon changed to rain. We did have a bit of accumulation on trees, but not enough to cause any power outages. We're supposed to have snow flurries this afternoon, but that's about it for this iteration of Winter in Winston-Salem. Click here to read or post responses to this week's journal entries [Top] |
|||
|
Tuesday, 8 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 8:50 - Hmm.
Another difference between the way women use language and the way men do.
"Worn out" (female) = "just broken in" (male). I
noticed this the other day when I brought the white load up from the
drier. I dumped them on the bed, preparatory to sorting Barbara's and my
stuff out. I always sit there on the bed with my underwear drawer open,
tossing my socks and underpants into the drawer as I separate them from
the pile. Barbara intercepted a pair of underpants and pointed out that
they had holes in them where the elastic waistband joins the body of the
garment. Being male, my attitude was, "Yeah? So what?"'
Men and women look at clothing differently. Men form close, long-lasting friendships with clothing, whereas women do not get emotionally attached to clothes. I know this sounds diametrically opposed to common perception, but I think it's true. Most guys would much rather wear something years old that they've worn hundreds of times before than risk wearing something new. I have stuff in my closet that I've never worn that's literally years old. I'll eventually get around to wearing it, but not until I've made friends with it. Like most guys, I'm suspicious of strangers until I get to know them, and that includes new clothes. I think Barbara sometimes tries to trick me by putting something new in the laundry, assuming that when I bring it back up from the drier I'll just put it in with the other clothes I wear regularly. To Barbara, one gray X-Large Tall LL Bean t-shirt is much like another. But I can tell the difference. Come to think of it, Barbara always leaves a couple of three-packs of new underpants in my drawer, but in the 18 years we've been married I've never opened one of those. Not even once. Since the number of underpants I have doesn't appear to be growing, and since the packages in the drawer look much too fresh to be 18 years old, I must assume that Barbara sometimes tosses some of my old buddies and replaces them with new guys. I'm still wondering where my sweatshirt from freshman year of college has gotten to. It's only 30 years old, and has stood up well other than a few minor tears, some color fading, a split seam or two, a bit of shrinkage, fraying edges, one missing sleeve, and so on. Barbara swears she never threw it out, which means it must be hiding around here somewhere.
Thanks to Wayne Ketner and Miguel Bazdresch for forwarding an interesting article from BBC News about asteroid 2001 YB5, a near-Earth-approaching object whose nearest approach is about half a million miles, or about twice the distance from Earth to Luna. In cosmic terms, that's nearly a direct hit. If that asteroid impacted at a typical orbital speed of about 70 kilometers/second, this 300 metre chunk of rock would cause devastation several orders of magnitude larger than the biggest thermonuclear device ever detonated. That's if it were to hit land, where the devastation would be localized, in the sense that the destruction would be largely limited to the continent it impacted. If it hit the ocean, as it would have a 70% chance of doing, the planet would probably suffer a year (or years) without a summer. If that doesn't sound so bad, consider that it would mean a billion or more deaths from starvation. And even that is an optimistic scenario. More likely, the water and dust vaporized by the impact would form an impenetrable cloud layer that would increase the albedo of Earth sufficiently to cause the onset of another ice age. And keep in mind that this is a very small object as such objects go. We have to do something about this problem. If we wait until the day we spot an object whose orbit intersects Earth, it will be too late to avoid the impact. We may have anything from a few months to a couple years to watch helplessly as doom approaches. Oh, I'm sure we'd try to do something. We might cobble up something with some mothballed hardware and send one guy on a one-way mission in a craft crammed with oxygen, water, fuel, and a large nuke, but it probably wouldn't work. We need real planetary defenses, and we need to start working on them now. Right now. A crash-priority project, like that for the Moon Landing, would give us reliable planetary defenses in less than a decade. An inestimable side benefit would be the resulting cheap access to space that would pay off big-time in many other endeavors. But we need to get started on it now. Few people realize just how high the likelihood is of an impact and just how catastrophic the results would be. The impact of a large object could literally eradicate humanity from the planet, and the likelihood of this happening in the relatively near future is high enough that everyone should be concerned. Perhaps not for themselves, but for their children, their grandchildren, and their grandchildrens' grandchildren. It might not happen for 50 years, or even 500, but the day will come when astronomers spot an object that is going to impact Earth. But before you get too comfortable thinking that 50 or 500 years is a long time, realize that it's just as likely that planet-killer could be spotted tomorrow as any other day in the next 500 years. We need to do something about it. We have the technology. All we need do is get out there while the object is still far away and give it a tiny shove. It doesn't even matter which direction we push it in. Up, down, to one side or the other, towards us, or away from us. Anything will do. But we need to be able to get out there while the object is still far enough away that a small delta-V will cause the object to miss Earth. Until, that is, we have orbital beam weapons that can direct enough energy at a remote object to affect its orbit. We can build the first solution with today's technology, and even the directed-energy solution isn't that far off. But we need the will to put that technology into play.
11:23 - I've had a couple of private emails asking if Earth's atmosphere wouldn't protect us against a planet-killer impact. The answer is no. The atmosphere would no more stop a large object traveling 70 km/s than a child's balloon would stop a rifle bullet. Depending upon the angle of incidence, the object would spend no more than a second or two in atmosphere dense enough to have any effect at all. And even that miniscule atmospheric braking wouldn't help at all, because all of the energy transferred from the object to the atmosphere would simply contribute to the destructive effect. The only way to protect the planet against such an impact is to modify the orbit of such an object to cause it to miss the planet.
This is very strange. I'm trying to install Mandrake 8.1, but I'm not having much luck. The first CD installs properly. It's the second and third CDs that I'm having the problem with. I downloaded the ISOs for all three and tried to burn them. The first CD burned normally. When I tried to use Nero to burn the second and third CDs, it gave me an error message about the block size not matching. Nero offered to fix that, so I told it to go ahead. The CDs then appeared to burn normally, but when I fed them to the Mandrake install process it didn't recognize them. So I tried inserting them in a Windows box, which also refused to recognize them. At that point, I assumed that the ISO images must have gotten screwed up during download or something, so I downloaded the ISOs for CDs 2 and 3 again, storing them in the \junk subdirectory of the main download directory. I tried burning them again, with the same results. So I finally downloaded the MD5 checksums and used Chris Madsen's FFV utility to run MD5 checksums against the ISO files I'd downloaded. According to Mandrake, these are the proper MD5 checksums for the three ISOs.
But when I ran FFV against the ISOs on my drive, it returned the following:
As you can see, the MD5 checksum for CD1 matches, but the checksums for the two copies of the other two files aren't even close, and in fact differ from each other. The checksum for CD2 should begin "587", but instead begins "838" and "6a1" on my two copies. Similarly, the MD5 checksum for CD3 should begin "598", but instead begins "b7e" and "68e" on my two copies. I can't remember for sure, but I think I downloaded the CD2 and CD3 ISOs from different mirror sites. What is going on here? Click here to read or post responses to this week's journal entries [Top] |
|||
|
Wednesday, 9 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 10:10 - I
now have all three Mandrake Linux 8.1 ISOs, and the MD5 hashes check for
all of them. Thanks to everyone who offered to burn CDs and send them to
me. I really appreciate your willingness to help.
After I wrote that entry yesterday morning, I went back to the Mandrake download page and downloaded the README file from one of the mirrors. In there, I noticed a warning about Netscape Navigator downloading in ASCII mode rather than binary mode by default. That got me to thinking. I'd downloaded the first ISO some time before I downloaded the other two. Could it be that I'd downloaded CD1 with IE and CD2 and CD3 with Mozilla? That was quite possible. I have so many machines and so many browsers installed around here that I couldn't remember what I'd used to download the various ISOs. But if Navigator uses ASCII mode by default for FTP transfers, I figured it was quite likely that Mozilla does as well. Convinced I'd discovered the problem, I fired up IE and started downloading CD2 and CD3. Just as CD3 had nearly completed, Brian Bilbrey called to tell me that he suspected my browser had downloaded the last two ISOs in ASCII mode. I told him what I just wrote in the paragraph above, and said that what really disturbed me was that I'd gotten two different MD5 hashes for each ISO. Even in ASCII mode, those hashes should have been the same. Oh, well. Brian suggested that if the MD5 hashes didn't check for the versions I'd just downloaded with IE, I should use my FTP client. After we got off the phone, I went back to my main system and ran Chris Madsen's Fast File Validator against the newly-downloaded ISOs. Not even close. So I did what I should have done originally. I fired up FTP Voyager on meepmeep, and started to download the ISOs for CD2 and CD3. That was harder than it should have been. I first pointed my FTP client at the North Carolina sourceforge.net mirror, which is the closest to me. The directory listing took forever to display, but I highlighted the two ISOs and started the download anyway. My FTP client was reporting throughput of 5 to 8 KB/s, which meant it would have taken all day to get the files. After watching that for a couple minutes to make sure it wouldn't improve, I cancelled out of it and tried a different mirror. No joy. I tried literally ten different mirror sites. The best of them yielded throughput about the same as the first one I'd tried, and the worst were so slow that the directory list didn't even appear within a minute or more. I was about to give up, figuring there was some kind of datastorm on the Internet or that RoadRunner was having local problems. Fortunately, I tried one last site. The one at the very bottom of the mirror list with no location given. As soon as I started the transfer, my FTP client showed throughput of 75 KB/s or so. That's not great, but it was ten times faster than the fastest I'd gotten on any other mirror, so I let it run. This morning, I remembered to check and found that the files were there. When I ran the Fast File Validator, all the MD5 hashes checked. So I'm now in good shape as far as Mandrake Linux 8.1. More on that as I get time to work with it.
I'm actually going to have two Linux boxes. The first one is already running Mandrake 8.1, and will be my Desktop Linux box. It's an Intel D845WNL motherboard with a Pentium 4/1.6 and 512 MB of SDRAM. It has an ATI video card, an 80 GB Seagate Barracuda IV ATA hard drive, and a Plextor PlexWriter CD burner in it. All standard components, with no SCSI or other stuff to complicate matters. The second Linux box is going to be a server. It'll sit behind my firewall serving files and functioning as my Linux server test-bed system. I'm going to use my current main Windows 2000 Professional system, thoth, for that box. I'll pull the current hard drive and replace it with a 180 GB SCSI Barracuda drive. There's already a DDS-3 tape drive in that box, which I may leave in there. If not, I'll replace it with a DDS-4 tape drive. There's also an ATAPI PlexWriter. As best I remember, the motherboard in that system is an Intel SE440BX2-V with a Slot 1 Pentium III/750 or /800 processor and 256 MB of SDRAM. That means I need a new Windows 2000 main system. I just happen to have an Antec SX640 case sitting here, with an Intel D845BG DDR motherboard sitting on top of it. I'm going to stick a Northwood-core Pentium 4 in that, add half a gig or so of PC2100 DDR-SDRAM, a big Seagate Barracuda ATA hard drive, a video card of some sort (probably a Matrox), and some sort of tape drive. All of that will have to wait until the book is finished, of course, but I'm looking forward to it.
Paul Jones from the Forsyth Astronomical Society emailed me yesterday to say that he'd gotten a new scope for Christmas and wanted to go up to Bullington to observe. Barbara was volunteering at SciWorks, but I told Paul that we'd probably be up. Barbara had a new course at her gym starting last night, so I told Paul we might not make it up there until 8:30 or thereabouts. I called Bonnie Richardson to see if she wanted to come up as well. She said she was going squirrel hunting in the afternoon but would be up later. Barbara and I finally got out of here about 8:15 and made it up to Bullington about 8:45. Paul and Bonnie were already there. That was unusual for us. Normally, we're the first ones there, but this time we had to drive in to the site with our lights off. As soon as we got out of the truck, it was obvious that we were going to have a good session. The night was extraordinarily clear, with the stars blazing down at us. A quick glance at Orion showed me stars down to 4th magnitude, before I was even dark-adapted. Unfortunately, clear skies often mean poor seeing, and last night was no exception. I realize that that statement is probably confusing to non-astronomers. Basically, clarity and seeing are two distinct aspects. A clear night means just that. Many stars are visible, and there's not much humidity or haze in the air. Stars jump out at you from a black background. Seeing, on the other hand, has to do with how stable the atmosphere is. On a night of good seeing, the atmosphere is stable and stars don't twinkle. On a night of poor seeing, stars twinkle constantly, and a bright white star through binoculars or a scope appears to be changing color constantly, from bright red to green to yellow to blue, and so on. During poor seeing, it's impossible to make out fine detail because the atmosphere is so turbulent. That makes it impossible to observe things for which fine detail is important--things like observing the planets, splitting close double stars, and so on. Bad seeing has less effect on dimmer objects like DSOs. (Deep Space Objects or Deep Sky Objects). Conversely, the best seeing around here often occurs during summer when there's lots of haze and relatively poor clarity. There are only a few nights a year around here when seeing and clarity are both excellent, and those are more precious than jewels. Last night, we had excellent clarity and mediocre seeing, so we concentrated on DSOs. Paul had his new Orion Short-Tube 80 refractor set up and was bagging Messier Objects when we got there. Paul is only his early thirties, but has 20 years of experience in astronomy, so he's very good at finding things. I wanted to try out my new 7X50 binoculars, so I wandered over to Paul and asked him if he wanted to have a contest. First guy to bag 25 Messier Objects wins. He agreed, and I started to use my binoculars to bag Messiers. I called out ten or so Messiers in the first couple minutes as I got them in my binocular--M31, M32, M33, M103, M52, M45, M42, M37, M36, M38--and Paul said, "Wait a minute. I thought you were going to use your scope, not binoculars. The bet is off." Heh, heh, heh. I can see why the Orion Short Tube 80 is so popular. They must sell them by the thousands. Bonnie had her ST80 up there as well. Basically, an ST80 is like half of a large binocular. They're relatively inexpensive (under $200 without mount), have a huge field of view (as telescopes go), and provide nice bright views of stuff like the Milky Way and star clusters. I prefer binoculars, but I can understand why so many people have fallen in love with these little scopes. As long as you recognize their limitations--their short focal length makes them less than ideal for planetary and lunar observing--they are excellent scopes. I don't know that I'd recommend one as a first (or only) scope, but they're a very good choice as a second, "grab-and-go" scope. They are also surprisingly competent for bagging DSOs. Obviously, an 80mm scope can't compete aperture-wise with the big Dobs, but the STs do quite a good job on the brighter DSOs. When Paul had his ST80 pointed at M42/43 in Orion, I suggested trying to bag M78 with it. We both laughed at that idea, because we thought M78 was too small and too dim to appear in an 80mm scope. But we pointed it towards where we knew M78 was, and sure enough a minute or so later Paul shouted that he had M78. It was a tiny, gray smudge that required averted vision to see, but he undoubtedly had M78 in the eyepiece. Barbara finally got a chance to use the 14mm Pentax XL eyepiece I got her as a Winter Solstice present. We used it with our 10" Dob and pointed the scope at M42, the Great Orion Nebula. At first, I couldn't believe my eyes. M42 filled the 3/4 degree field of view with bright nebulosity. At first, I honestly thought there was a cloud in the field of view. The four brightest stars of the Trapezium split cleanly at 90X, and at one point I got a glimpse of the fifth star. That really surprised me, because seeing E (and particularly seeing F) normally requires much higher power and very steady seeing. The problem with E and F is that they're a goodly distance from the four main Trapezium stars and tend to disappear into the background nebulosity. I never did get E again during the evening, but even getting that one glimpse of it made my night. The 14mm Pentax XL eyepiece is indeed everything it's reputed to be. It's a world-class eyepiece, and the views through it are extraordinary. The 20mm of eye relief makes it a very comfortable eyepiece to use. I'm looking forward to getting some serious observing time in with this eyepiece. We didn't spend long at Bullington last night. Although the temperature was just below freezing, none of us were feeling particularly cold at first. But by 10:00 or so, Paul and I were both starting to feel chilled. He announced that he was going to pack up, and Bonnie said she was as well. So Barbara and I decided to make a short night of it as well. We got out of there around 10:15 and home by 10:45. It was a short but enjoyable evening. Click here to read or post responses to this week's journal entries [Top] |
|||
|
Thursday, 10 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 9:31 - I
burned CD2 and CD3 for Mandrake 8.1 yesterday, which took longer than it
should have. I'm out of good CD-R blanks, so I fell back on the Spindle
That Will Not Die of 4X CD-R blanks I got when I bought my Smart &
Friendly CD burner some years ago. I hate these blanks, but I keep them
around for emergencies. As a result, the original spindle of 100 still has
probably 50 discs left. Which is very strange, considering it feels like
I've used about a thousand of those discs already.
At one point, I was ready to pitch the remaining discs because even the Smart & Friendly drive choked on them frequently. Unfortunately, the Plextor PlexWriters are so good that they'll use almost anything, including these crappy blanks. Even the Plextors burn an occasional coaster with these blanks, but they're better than nothing. I need to order another 100 spindle of Taiyo Yuden 24X blanks. Once I got CD2 and CD3 burned, I took the advice several people had given me, and did a kitchen-sink install of Mandrake 8.1. At this point, I have a functioning Mandrake installation. I tested it by using Navigator (ugh) to hit a couple of web sites, and by playing a game of Solitaire. And, boy, is Solitaire ugly. I mean UGLY. But at least I have a functioning Linux workstation now. I don't have time to do much with it yet, but it's there when I have time to get to it.
Yet another search engine has fallen by the way. Northern Light announced that it will no longer offer its general search engine to the public, having decided to concentrate on corporate accounts. It's been an interesting progression. In the early days, we all used Lycos because it was the only thing available. Then DEC opened the AltaVista site, and it was so much better than Lycos that searchers abandoned Lycos in droves. Then Northern Light came along, and many of us abandoned AltaVista. Then Google arrived with a splash, and a lot of people started using it exclusively. I'm one of them. I haven't used anything other than Google for general searches for the last year or two. At first, that was because Google returned better results than any other search engine. As of late, that's also because Google is the only mainstream search engine that still returns results ranked according to the searcher's criteria rather than according to who has paid them for placements. Of course, the complete failure of banner advertising has hurt all of these companies badly. What I don't understand is why Google hasn't instituted a subscription plan. If Google charged a reasonable annual fee, say $24, for unlimited access to their search engine, a lot of people would subscribe, especially given how poor the "free" alternatives are. That $24 equates to a lot of banner ads. Some sites are now getting as little as $0.10 revenue per 1,000 banner impressions, so they'd have to sell a lot of banner ads to equal the $24 per person annual revenue. Google could continue their free ad-supported service, but limit its scope. Subscribers would have access to the entire database, access to third-party databases, enhanced search features, freedom from banner ads and pop-unders, and so on.
Dvorak is at it again, this time with an article entitled The Nine Assassins of Broadband. Does this guy live in the real world? If you believe him, broadband doesn't work, it's not good for anything, and no one wants it anyway. Give me a break. No one I know who has broadband would ever voluntarily return to dial-up, and everyone I know who has only dialup would kill for broadband. And that's not just techies, either. I'm talking about regular people I know--from students to stay-at-home moms to grandparents. Dvorak is an idiot. Click here to read or post responses to this week's journal entries [Top] |
|||
|
Friday, 11 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 9:40 - Lisa
Beamer, the widow of Todd Morgan Beamer, one of the heroes of United
Flight 93, gave birth to a daughter yesterday. Morgan Kay Beamer weighs 7
pounds and is 21 inches tall. She will never know her father, thanks to
those bastards.
I At first, it seemed as though it would be easy enough. I fired up Konqueror and went over to the Ximian website. There I found the main page for Evolution, and clicked on the download link. I was presented with an FTP directory listing that included many distributions of Linux. There wasn't a Mandrake 8.1 directory, so I chose Mandrake 8.0 as the next best alternative. It does seem odd, though, that Ximian doesn't have a Mandrake 8.1 directory almost four months after 8.1 shipped. Once I changed to the Mandrake 8.0 directory, I got a listing that included many files, most of which were apparently unrelated to Evolution. But there was one file that sounded like a reasonable choice, named evolution-1.0-ximian.3.i586.rpm. So I clicked on that one to get it. I was quite pleased at first. Rather than just asking me where to store the downloaded file, Linux gave me the option of installing it on the fly. So I chose that option, and at first things appeared to proceed normally. Unfortunately, I was soon disabused of my optimism. The installer popped up a dialog that told me it couldn't complete the installation because several required libraries weren't present. I couldn't see any alternative but going out and locating those libraries one-by-one, downloading them, and installing them. But it seemed there had to be an easier way, so I emailed Brian and Greg. They told me that I could indeed do it the hard way, but that it'd be harder than I expected, because each of those dependencies might have dependencies of its own. That way lay insanity, so I decided to try some of the alternatives first. Brian suggested visiting this page, which would grab all of Ximian Gnome for me. I tried that, and I'm sure it would have worked, except for one minor problem. Those instructions assumed that the machine was connected directly to the Internet, whereas my Linux box is accessing the Internet through my WinGate proxy server. I'm sure there was probably some way at the command line to tell it to use the proxy server, but I didn't know what that might be. Plan B had failed, so it was on to Plan C. Brian or Greg had mentioned that because an early beta of Evolution had already been installed from CD, I might be able to use the MandrakeUpdate utility to update all of Mandrake, including Evolution. That utility worked fine to update Mandrake itself, but it left me with Evolution 0.13 installed. So much for Plan C. So I went to Plan D, which was to install Ximian's Red Carpet updating utility. I downloaded and installed the RPM for it without problems, but when I fired it up it turned out to be useless for my purposes. Red Carpet has "channels" that you can subscribe to. By default, I wasn't subscribed to any channels, so I downloaded the available channels, planning to subscribe to the Ximian channel and grab Evolution 1.0 that way. No joy. The only channel available was Mandrake 8.1. I got an error message during the download, telling me that the Ximian channel was unavailable. I'm not sure if that's because they don't have a channel for Mandrake 8.1, or if it was caused by their ftp server, which is pathetically slow. And I mean pathetically. While downloading the rpm's earlier, I was getting throughput of 3.5 KB/s on my cable connection. At first, I thought perhaps RoadRunner or the Internet in general was the problem, but everything else was running at normal speed, so it was in fact simply horrendously slow throughput from the Ximian server. The whole time I was going through this nightmare, I kept thinking of how easy the same process is in the Windows world. You download one file, unzip it if necessary, and run Setup.exe. If a Windows application has a DLL dependency, that DLL is either included in the distribution file, or the installation routine goes an gets it by itself. Linux should work the same way. Don't TELL me there are necessary files missing. GO AND GET those files for me, and install them. Don't leave me hanging. Assume you want a house. With Windows, you go out and look at houses, choose the one you want, buy it, and move in. The lights work, the toilets flush, and the furnace keeps the place warm. With Linux, you don't have to buy the house. Someone will give it to you. But first, you have to go out and choose among piles of lumber and stacks of concrete blocks. You can have any house you want, but you have to build it yourself. And, after you're done, the toilets don't flush and the furnace doesn't heat the place, because you haven't yet built the sewage treatment plant, drilled the natural gas well, built the power plant, or laid the pipelines. I am completely disgusted with Linux and Ximian. But this morning I get the following emails from Brian:
and
Arrrrrgggggghhhhh. All I wanted to do was install one stinking application under Linux. If even Linux gurus have to struggle to do something as simple as downloading and installing an application, Linux is not for me. Or at least not yet. I do wish there were a Windows version of Evolution, though. From what I saw, it looks like a very nice alternative to Outlook. I've already blown away the Mandrake 8.1 installation on the Pentium 4 box, and have Windows XP Professional installing on it as I write this. I'll use that to do screenshots for the book. Once I'm done with that, I'll load Windows 2000 on that box. Windows 2000 is fast, stable, and easy to use. I'll look at Linux in another year or so, to see if it's gotten any less user-hostile. Don't get me wrong. I'm not abandoning Linux because I spent six frustrating hours failing to get one application installed. I'm abandoning Linux because the problem was not me. If Brian or Greg had emailed me at the end of that six hours and said, "Bob, there's an easy way to do that. It takes five minutes", I'd have written off the six hours to a learning experience. But that's not what happened. I'm not going to cripple myself by adopting an operating system that makes even easy things hard, even for experienced users. In fact, I am writing off that six hours to a learning experience. Just not the kind of learning experience I expected.
12:33 - Hmmm. Bilbrey says I'm a wimp and a wuss, and he's disappointed in me for giving up so easily. So we have the following exchange of emails:
Not frustrated. Disgusted. I'm prepared to deal with frustration. I expected it. But again, we're talking about *installing an application*. That's not equivalent to installing a bidet with sidecar. That's equivalent to adding another fork to the silverware drawer, or it should be. Silly me. With Windows, it would have been a matter of (a) open drawer, (b) drop fork into drawer, and (c) close drawer. With Linux, it appears to be a matter of tearing out the entire damned kitchen and remodeling it. Or perhaps I'm being unfair. But I don't think so. You say that this is a five-minute solution, and perhaps it is. But if someone who was a Linux expert and had no experience with Windows called or emailed me for help with installing an email client, any email client, I could have walked him through the process in about a minute. Doesn't matter which version of Windows, or whose email client. Doesn't matter whether he's directly connected to the Internet or using a proxy server. Someone else has pointed out that Windows apps have dependencies as well, and that something designed to run on NT4 might not run on XP. Others have pointed out that most Linux apps install easily and that Ximian is simply user-vicious. Several people have commented that Evolution was probably the worst single app I could possibly have chosen for my first effort. Fair enough. Okay. You're right. I don't give up easily. I'm sure I'd have made it through SEAL training, not because I'm tough (I'm actually a wimp) but because I'm used to winning. I mean, when I decided at age 15 to learn to play tennis, I spent most of a summer hitting balls off a backboard before I stepped onto a tennis court. So I've re-installed Mandrake 8.1 and am now sucking down Evolution 1.0 via the French site you mentioned in an earlier mail. Oops. A little box just popped up to say "An error occurred while fetching file libGConf1-1.0.7-2mdk.i568.rpm" Hmm. The only option is Skip, so I'll do that and see if it gets the next one, even though I realize that it's useless to proceed without all of them. Ah, I see it just blows up with an "Installation Failed" message, so I'm not sure why it even gave me the Skip dialog. Okay, back to the lists, where I see that Evolution 0.13 and 1.0 are both showing as "Installed" and neither is showing as "Installable". Okay. Remove both instances of Evolution on the "Installed" list. They don't show back up on "Installable" so I did a refresh. We'll see what happens. And after half an hour the little progress bar was still boucing back and forth, with no apparent progress. I tried to exit the application, but that didn't work, so I simply rebooted the system. I've been looking at a message for about 10 minutes now--"Shutting down interface eth0:" Power reset time... Hmm. I see that Linux is going through what appears to be an fsck, so I'm going to bag this copy and do YALI. This'll be only the fifth time I've installed Linux in the last 24 hours, so that's not so bad. There are a lot of nice things here. At first glance, for example, the Konquerer browser appears to be a very nice product. Nicer than Mozilla, and possibly nicer than Opera, although I haven't tried installing it under Linux yet. There are also a lot of other nice things about Linux and Linux apps, which I didn't mean to minimize in my rant. As someone said, I was complaining about installing an application that wasn't officially supported by my OS version. So I'm going through the Linux installation procedure yet again, this time accepting all the default options. I hate it when the other kids make fun of me. I still say that Linux isn't ready for Aunt Minnie, but then I'm not Aunt Minnie. I'm going to give this Evolution installation another shot (okay, maybe three or four more shots) because I really want Evolution as my PIM/mailer.
Maybe it's just me, but the spam type I most resent is that which forges my own address as the sender. Here's an example of one of those.
Click here to read or post responses to this week's journal entries [Top] |
|||
|
Saturday, 12 January 2002 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 10:30 - Something
isn't right. If you haven't read John
Dominik's journal page lately, check out this week's entries. Here we
have a smart guy who works hard, does his job--including stuff that goes
beyond his job definition--and still he may be about to be laid off. That
wouldn't be so strange if John were the most recently hired of 20
programmers or some other similarly interchangeable cog, but instead he's
the only guy in his office who does what he does. How long can a company
that lives and dies with its PCs and servers survive without a
systems/network administrator? Something isn't right.
This is particularly heartbreaking, because John was just beginning to get back on his feet after losing his last job, also through no fault of his own. He and his wife were planning to buy their first house, but that's all on hold now because his employment situation is so uncertain. I suppose he should be thankful that this problem cropped up when it did rather than a few months from now when he was committed to paying a mortgage, but I'm sure John doesn't feel thankful that his job is again in question. Maybe John is just unlucky. Certainly any particular company can encounter difficulties. But I don't think that's the explanation in John's case. I've heard this story too often lately, and it seems to me that despite what the government would have us believe, we're not just in a minor economic slowdown. We're in a full-fledged depression, and that depression seems to be affecting the hi-tech industry more than most. Or perhaps I should say earlier than most. The dot-com implosion certainly had a lot to do with it. When that wave was riding high, there were more technical jobs available than there were qualified people to fill them. Now that things are bottoming out in hi-tech, it seems that there are ten technical people available for every one job advertised. I'm seeing it happen locally as well. People with killer resumes are now flipping burgers or delivering pizza, in some cases literally. Companies that were high-flyers a year or two ago are now laying off staff in droves. People I know who have extraordinary high-tech skills now feel lucky to have a job, any job. Other than Microsoft, which has something like $34 billion cash in the bank, no high-tech company I know of is hiring in any numbers. Even those in the best shape are reducing their work forces by attrition. Of course, all of this is temporary, but that's cold comfort to people who are unemployed or may soon become so.
And here's a message I sent to one of the mailing lists I belong to: I always wonder what people mean by the phrase "steep learning curve". Time is conventionally mapped to the x-axis, which means a product with a steep learning curve can be learned quickly and easily, i.e. a large amount of learning (y-axis) per unit of time (x-axis). If a product is difficult and time-consuming to learn, it has a shallow learning curve. The earliest use I've seen of the term "steep learning curve" was in some pioneering time-motion studies from before WWII, in which the term was used correctly. Some people still use it correctly, but it appears they're in the small minority these days. I wonder if the meaning of the phrase was corrupted because steep hills are harder to climb than are shallow hills. I think there's a word that describes the corruption of a word or phrase to the point that it assumes exactly the opposite meaning of what it states. For example, using "I could care less" to mean "I couldn't care less." I wish I could remember that word. I've about given up the struggle for restoring the original and proper meaning of "steep learning curve", though. Even Jerry Pournelle, who is a very careful writer, uses that phrase to mean "difficult to learn". I've called him on it several times when I was doing sanity checks on his Byte columns, and each time his response is, "Yes, yes, I know, but..." So this once meaningful phrase now has zero semantic content. When I hear or see it used, I'm never sure, except perhaps from context, what the person means. The probability is high that he means "difficult to learn" but perhaps 10% of the time the person will be using the phrase correctly. Back in pre-Web BBS days, there was a GT echo called "English as it is properly spoke", where issues like this were debated. I wonder if there is a similar newsgroup or mailing list.
No playing with Linux for me for the next week or two. I have to finish my lecture for the astronomy club this weekend, and then I absolutely must get the final two chapters for PCHIAN done and sent off. That means things will be sparse around here, too. I have 30 "real" messages in my inbox to deal with, and then I'm going to disappear for a while while I get my work done.
Click here to read or post responses to this week's journal entries [Top] |
|||
|
[Last Week] [Monday] [Tuesday] [Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next Week] |
|||
| 10:40 - I'm
getting lots of advice about which Linux distribution to use, and the
common thread is that I should try several and choose the one I prefer. My
problem with that advice is that I'm so ignorant about Linux that I'm not
yet entitled to have an opinion. Roland Dobbins, Brian Bilbrey, and others
whose opinions I respect tell me that Mandrake is not really mainstream,
and that I'd do better to use Red Hat 7.2 initially, and later transition
to Debian. Apparently, Red Hat is dominant enough that it pretty much
defines the mainstream, and Debian's apt get functionality indeed does
sound very nice.
So I started to download Red Hat 7.2 yesterday afternoon. Looking over at meepmeep right now, I see that my ftp client is on its second attempt to download Disc 1, and is 558 MB into the download, but showing only 5.95 KB/s, which isn't much better than dialup. Fortunately, my ftp client is relentless. Red Hat is just down the road from us. I could have driven down there and picked up Red Hat 7.2 in less time than it would take to download it even if the cable modem was providing its usual throughput. I thought the Ximian ftp server was slow, but the Red Hat server makes the Ximian server look like the proverbial firehose. Are there mirrors of the Red Hat server? If so, I think I'll try one.
Bloomberg posts an interesting article, entitled WTO Seen Ruling U.S. Tax Law Illegal; $4 Billion EU Duties Loom. As far as I can see, the European Union is taking the US to task for doing exactly what European countries have been doing all along. Now that the US is doing it, though, it's unfair competition. Hmmm. I always thought it was interesting that if, say, a Brit comes to the US to work, he pays US taxes but not UK taxes on the income he earns in the US, while if I go to Britain to work, I pay both UK taxes and US taxes on the income I earn in Britain. The US is one of only a handful of countries that attempts to levy income taxes on moneys earned outside their jurisdictions. As far as I can see from the Bloomberg article, the US has simply started to level the playing field. If the EU wants to start a tariff war on that basis, fine. They'll lose. As I've been saying all along, the US needs to withdraw from international entanglements such as NATO, the United Nations, and WTO, and pursue our own course unilaterally. The US should be solely concerned with the interests of American citizens and companies, without regard to the interests of anyone else. I'd suggest a good starting point would be to eliminate Most Favored Nation trading status for all countries other than Canada, Mexico, and Great Britain. Then we can add back countries that have historically been friendly to the US, assuming that they are willing to defer to our economic, political, and military interests. Those who are willing to be a Friend of Rome will prosper. Those that are not will not. The French now describe the US as a "Hyper-Power", but even that does not go far enough. The US is now not only the top spender on defense, but spends more than the rest of the top ten combined. If this trend continues, as it almost certainly will, the US will soon spend more on defense than all other countries on the planet combined. Militarily, not only is there no country which can contend with the US, there is no combination of countries that can do so. The US is secure in its ability to defend itself, and to punish any country or coalition of countries that attempts to use military force to counter US interests. It's pointless being a 900-pound gorilla if you don't take advantage of that. Pournelle and I have concluded independently that empire is now inevitable, and that if we're going to have empire we'd just as soon have an efficient empire as an inept one. That will require some attitude adjustment on the part of US citizens and politicians, primarily a change from our desire to be liked to a desire to be feared. This is where we're heading. Expect the trend to continue. Click here to read or post responses to this week's journal entries [Top] |
|||
