© |
Daily notes and commentary -- Week 23* Link to: last modified 01:50 GMT+2 on 14.06.1999
Hi, welcome to this week's daybook page.
Associated links:
See also the Daynotes index. |
© |
Monday 07.06Monday vanished somewhat in a flurry of extraneous detail, not to mention the intense focus of editing chapters.
Bob Thompson mentions a quote on
his Monday
page
|
© |
Tuesday 08.06A bit late with yesterday's posting. But good things are generally timeless... :) Will in fact be even a bit later. There are other things to do demanding my attention now, and I note some connectivity problems on the Net somewhere between here and US sites. Timing out in various places as the morning progresses here.
(later) Bob points out that the first quote attribution is spurious, though common -- the quote is he says by Ambrose Bierce. It could be either, but I've not found any reference to it in my usual places. I've posted the thread on the wiki. (later still) Large companies get bad habits. Inquiring about the status of my overdue invoices turned up the annoying fact that the company's Stockholm office, where major administrative and financial functions are concentrated, this year have a new "procedure" for incoming invoices. One should here also be aware that the invoices first wind a somewhat brownian path through the local offices for payment authorization, before being sent up to central HQ for actual entry into the payment system. Turns out that the payment due date they run on is strictly 30 days , all right, but not from the invoice date, nor even from the date of receipt -- but from the date it is entered into the invoice payments system. So much for my "30 days"... I should be so lucky to use the same procedure on my bills. "What? Overdue? No sorry, the bill will be paid first next month, because I only got round to entering it into my ledger yesterday, and well you know, payment around here is strictly 30 days from then." Anyway, my local invoicing contact promised to expedite matters through some internal channels. We'll see. Overdue charges will be billed, no matter what.
|
© |
Wednesday 09.06Cackle, gibber, they're coming to take me away, haha haha haha... No sooner had I read Tom Syroid's comments about multi-OS boot problems, than I of course send him a message about Partition Magic (which he also praises highly -- yes indeedee, highly recommended) and the in v4 bundled BootMagic. And having done that, I ponder, why have I not installed that myself yet? I've been meaning to, but the OS Loader and boot.ini setup from the NT install just kind of keeps rolling on. So, ok, in a free moment (of which I have so many <g> -- though they tend to be measurable in multitasked CPU cycles rather than seconds, and as such get shorter by the year <bg>), I whipped out my PM-CD and performed the appropriate magic invocations. Trundle-trundle... Everything installs nicely, and the boot diskette is also made. Then I fire up the configure program and get an odd display.where the only OS listed is Windows NT -- Hmmm? I currently run Win95 on C and NT on D, so that's not right. Following instructions, I select Add and then hit the button Advanced to see all partitions. Curiouser and curiouser... Every one except the first is listed as unknown, and unlike what it says in the manual, volume name and other extra info is not listed at all. Still, figuring out the correct partitions was not difficult, but when it came around to testing time, sorry, nope, no NT -- the only thing that would boot was the default first partition, i.e. Win95. Weird. Well, I wasted an hour doing this, doing that, and looking for some clue, but got nowhere, so I'm back to boot.ini. I guess that's what I get for recommending something I'd not installed myself yet. Karma works in mysterious ways...
The web is a wonderful thing -- the above comments about BootMagic led to some quick replies, posted on wiki. Join the discussion...
|
© |
Thursday 10.06Putting in some time to fight creeping linkrot. On the whole I feel I run pretty link-tight sites, but what with hosting changes and the forced upper case due to restoring a backup onto a Windows partition, I have to run a few checks now and then even on supposedly stable in-site links. There are a few puzzling entries in the website error log, however, and I can't quite determine from where they come. It can be either an external request, or one of my links somewhere. Even running automatic link checkers such as Xenu is not 100% foolproof, I find, although it is a very good utility. The fault lies not so much in Xenu, as the fact that I have numerous cross-referencing sites, and running complete verification passes with external links turned on is somewhat time consuming. Reading Bob Thompson's remarks this week about the trials of moving a mixed-case-linked site from an NT server to a Unix one, I can easily visualize what he has been going through. And as I noted earlier, the MS assurance that while case is ignored, it is preserved, is not worth much, nor even properly speaking always true -- it is preserved only if automatc case adjustment is disabled, and the filename does not comply with DOS 8+3 format (i.e. forces an alias). It is a shame, reading about the problems.encountered by those using Frontpage. FP having offered the potential of being a good web publishing application, sadly seems to let the user down in a number of ungood ways. (I had actually wondered why site size for FP hosting tended to cluster at 20 Mb, but thought this was a just minimum to make the extra overhead for the FP server side extensions worthwhile. So, it may instead represent a choke size for FP itself.) The absurdity of FP forcing a re-publishing of all site files for no apparent reason is obvious. The inadvisability of ftp'ing of files to an FP site, assuming that the host will even let you, is another. I had contemplated putting up FP-hosting on one of my sites, if only to properly test FP2000, but lately I am considering doing any FP webpublishing only over a local network, not a public website. We'll see -- it's nothing I'll be doing soon in any case. There are other projects that have priority at present.
|
© |
Friday 11.06This was one of those days when I wanted to launch multiple instances of myself -- there were numerous disparate demands on my attention and presence, and a single instance was simply not adequate. Hence the delayed update of this page (Saturday). End-of-school-year was one demand. Two kids, two different schools, so Isabel took one and I the other. Then of couse the kids were free as of lunch time, with further demands on my presence. Heck, they need the attention, the way I've been locked to the keyboard lately. (Say after me kids, Deadline. Dead-line. ...) Especially Therese "wants her Daddy" to be in on a number of projects of her own. Play a (non-computer) game, woodwork, painting, read a story, help tidy up, ... Edward is much more likely to simply as if he can be on another computer. A lot of fast email exchanges marked the later part of the day and evening, as Tom Syroid and I try to sync the style of our chapters with the wishes of our editor at O'Reilly. This is the second major milestone in the project, submission and approval of the two first chapters. The first milestone of course was having the proposal accepted and contract signed. This time round, its all in the details (and delays) -- so much more when we are still learning the ORA way of doing things. There has been some confusion about how to correctly apply the style guidelines, with new discoveries daily -- and a fair bit of inconsistency between different guideline sources and the published books that could stand as examples. I would call this equal parts following the signposts and hacking one's way through the underbrush.
|
© |
Saturday 12.06
Ah, a quiet and peaceful summer-holiday Saturday. Back to the keyboard - typetypetype :) Actually, I'm not typing that much today. I ended up in reading mode, between other chores.
Some days back, I downloaded the essay In the Beginning was the Command
Line, by Neal Stephenson -- something long enough and interesting
enough that it takes significant time to read and digest. (I provide it here
as a Zip
file Even assuming that html serving changes radically, I have the wiki source and can adapt the rendering as required, with no change to the notes at all. This is quite unlike traditional PIM applications, database files, or even word processors, where you are at the mercy of changing proprietary file formats that put you at risk of one day, after an upgrade, finding that you can no longer read your older notes.
|
© |
Sunday 13.06* A couple of comments from the NT News list that drops into my mailbox from time to time...
I keep finding new wrinkles with Outlook (as does Tom). Today I went to close one of my other Internet programs and it threw up a warning at me saying that there were active TCP/IP connections present and dropping them might cause instabilities. Say what? I checked around, but everything seemed normal and offline as far as I could tell. Turned out that Outlook was somehow holding a socket "open" to nowhere in particular. Closing Outlook removed that, and there were no further objections from the other program. The other day I was going to rename a folder and this was denied. After much head scratching I finally concluded this unlikely connection: I had edited a file in that folder using Word, and Word had also been used to write an email. Outlook was still open, but Word had been closed. Outlook still asserted some kind of indirect lock on the folder Word had accessed. When I closed Outlook, I could access and change the folder normally. From a few limited experiments, I have verified that all Office components show this interlocking aspect to some extent, and will deny access to folders and files previously held by other components. The most trivial is that it is impossible to delete orphaned Word temp files (say left over from a crashed session days or months ago, even in totally unreleated locations to the current session) unless Word is not running.
|
All original material Copyright 1999 Bo Leuf. |