I received notice this morning from eBay:

We received a report about a message you sent to another eBay member through our Email Forwarding System. The message violates the Misuse of eBay Email Forwarding System policy. We want to let you know about the report and invite you to learn more about communication between sellers and buyers. To learn more about the Email Forwarding System guidelines, please go to:

http://pages.ebay.com/help/policies/rfe-unwelcome-email-misuse.html

We're taking a neutral position regarding the report we received, but if we continue to receive similar reports, we'll have to investigate. Policy violations can result in a formal warning, a temporary suspension, or an indefinite suspension.

If you have concerns related to this matter, you can contact us by going to:

http://pages.ebay.com/help/contact_us/_base/index_selection.html

Well, I'd like to know about what message this complaint was levelled. But, naturally (this being eBay), there's no appropriate option at index_selection.html, and the best fitting options require that in one field I provide a relevant item number or user ID about whom I'm complaining. My own user ID is rejected from this field.

Over the years, eBay, like many other corporations, has modified its interface and protocols to make them dumber in ways that specifically increase the difficulty of confronting it with responsibility.

eBay was founded by Pierre Omidyar, whose user ID is pierre. So I entered that user ID in the field, and it was accepted. Doubtless that, if others do likewise, then the software will be tweaked to prevent it.

Years ago, core computing tasks were performed on shared mainframe computers, with individual users assigned terminal devices to communicate with the mainframe computer. Some of the terminals were smart, and able to enhance the interaction. Notably, the CTC Datapoint 2200 was in fact itself a programmable computer (in production five years before Steve Wozniak's Apple I), and was the direct ancestor of the x86 computers of to-day. But few of smart terminals themselves ran any code other than to provide interface for communication with the mainframe. (And the dumb terminals ran no applications.) There were efforts to get the general population using mainframes by way of terminals located in their homes, but these efforts enjoyed limited success.

Then the idea of personal computers caught hold, so that a large share of the population indeed computed at home or in the office. But the computing itself was primarily done at basically the same physical location as was the user. It was possible to add some communications hardware to the computer, and then use it as a terminal device, but most of the tasks that had previously been performed on a mainframe were now being performed locally.

When the 'Net came into wider use, some people started having the thought that perhaps it would be an advancement if principal computing tasks were moved onto the Internet, which is to say onto serving computers that were available by way of the 'Net. Unsurprisingly, I see this as a return to an earlier, previously unpopular model.

Now, sometimes, changes in an infrastructure can breathe new life into essentially older technologies; and one shouldn't reject this idea of moving back to locating core computing on remote machines simply because we had previously reduced its relative use. But I find it a signally unappealing idea, because it removes the independence of personal computing. I very much like the fact that I can do everything without communicating except communication itself. I have local applications for text processing and for type-setting, for multimedia generation, for mathematical analysis, and for programming. For these things, I don't have to rely upon a connection to the 'Net nor upon someone's server.

While there are some tasks that might be better performed by a network of distributed service, there is no particular reason for handing responsibility to such a network for mundane tasks that users could easily be performing with local equipment. And the introduction of the opaque buzzword cloud to refer to distributed service on the Internet does nothing but get my back up.

Anyway, I was prompted to ventilate by this story:

T-Mobile Sidekick users have had things such as contact information and photos stored on the cloud, which in this case is to say some servers that T-Mobile has leased from a division of Microsoft, which division is aptly named Microsoft/Danger. Well, the servers weren't properly backed-up, they crashed, and most or all that they held is just … gone.

[Up-Date (2009:10/15): Microsoft has now largely reversed itself, declaring we have recovered most, if not all, customer data for those Sidekick customers whose data was affected by the recent outage. We rebuilt the system component by component, recovering data along the way.]

Some years ago, Go Daddy added a back-order service whereby a domain already registered by another party would be monitored and, should its registeration lapse, the domain would be registered in the name of the purchaser of the service.

Some time after that, Go Daddy added an auction service. And, when registration lapses for which Go Daddy is the registrar, then duiring the grace period (42 days in the case of Go Daddy) Go Daddy itself puts the domain up for auction; the auction ends well before the grace period, and the auction results are cancelled if the prior registrant renews before the end of the grace period.

A registrant seeking to have a domain appraised might simply let the registration lapse, watch the auction, and then register before the end of the grace period. A late registration requires a higher fee, but that difference could be viewed as the cost of appraisal.

Now. here's where it gets ugly. Go Daddy holds such auctions even if there is a prior back-order. They hold the auctions even if the domain had a different registrar when the back-order was placed, but then switched to Go Daddy. They hold the auctions even if the back-order was placed before they had an auction service. If the domain should be bid to anything above an opening bid of US$10, the purchaser of the back-order must either pay more or let the domain go to some other party.

A Go Daddy back-order on a domain is worse than useless to its buyer if the domain may be expected to be registered with Go Daddy at the time whenever registration lapses. If the domain is sufficiently attractive that a back-order would be useful without an auction, then there will be competitive bidding in an auction.

Now, I'm sure that Go Daddy sent a notice that the old back-orders were going to be subjected to the new protocol, and that refunds were offered. But few if any customers would have understood the implications of a change, otherwise there would have been a lawsuit that Go Daddy would have lost, as simple refunds wouldn't have covered the economic loss avoidably being placed on these customers.

Last night, I finished the clean-up of a LAΤΕΧ version of my paper on incomplete preferences. From remarks by a person more knowledgeable about ΤΕΧ than I, it seemed that my best option in dealing with the under-sized angle brackets was to just fall back to using only parentheses, square brackets, and braces for taller delimiters. And most width problems were resolved by expressing formulæ over more lines. Unfortunately, these changes leave the formulæ harder to read than in the original.

This after-noon, I completed the submission process to one of the two specialized journals recommended by the advising editor who rejected it at the previous journal to which I submitted it. The submission process for this latest journal required that I name the other journals to which I'd submitted the paper. As simultaneous submissions are disallowed, basically they were asking for a list of which journals has rejected the paper. I gave it. (I didn't tell them that the third had been suggested by the second, nor that theirs had been suggested by the fourth.)

Anyway, I'm back to waiting for a response.

Large parts of Sunday and of Monday were given-over to producing a LAΤΕΧ version of my paper on incomplete preferences. Some part of to-day will be spent trying to complete that conversion.

Each of the two journals suggested by the advising editor who last rejected my paper requires submissions to be either LAΤΕΧ or in the form of a Microsoft Word file with formulæ encoded for its equation editor.

The original is an ODT, whose equations are encoded for the OpenOffice formula editor. The OpenOffice software can export a Word .DOC, but the formula would be rendered as text (rather than for the equation editor), and pretty badly at that. I used Writer2LaTeX, a plug-in for OpenOffice, to create a first-pass version of LAΤΕΧ source for my paper. A great deal of formatting went by the way-side, but the formulæ themselves seem to have come through the process mostly intact. The worst glitch so far was that U+2280 (⊀) was translated to \nsucc (⊁). (I've contacted the developer.)

I've nested the formulæ within more appropriate mark-up, and wrestled the rest of the mark-up of the paper into pretty good shape. The most glaring problems that I have right now are with the formulæ; those angle-brackets that should be rendered quite large are not, and some of the formulæ are simply too wide. And I still need to walk through the paper to make sure that my copying-and-pasting didn't go south anywhere.

---

This conversion of my paper represents the first time that I have worked much with LAΤΕΧ beyond creäting bald formulæ.

When I first started doing word-processing of more than plain-text files, LAΤΕΧ itself didn't yet exist, the ΤΕΧ system (on which LAΤΕΧ is founded) was still quite new, and I had personal connections to AT&T, such that I learned and used an older system, troff, which had been developed at AT&T (to justify the development of Unix). I had (and somewhere still have) a porting of troff to MS-DOS, could get pretty much any desired result with it (unlike most troff users, I knew its mark-up pretty thoroughly, and didn't rely on macros other than those that I'd written myself), and saw little reason to learn ΤΕΧ or LAΤΕΧ. When I did migrate from troff, it was to WYSIWYG programs.[1]

---

[1] There was an attempt, called Scientific Word, at something like a WYSIWYG interface for LAΤΕΧ; but, at the time that I investigated Scientific Word, it was ghastly. Its installation routine took hours, and, if the system already had a lot of installed fonts, would fail at the very end, in a way that its own programmers could not diagnose. And, if successfully installed, Scientific Word wouldn't produce decent LAΤΕΧ source anyway.

If you're actually trying to install another version of Firefox, then click on the Firefox tag, as there may be an entry on that other version.

Firefox 3.5.2 has been released. I imagine that someone will soon provide an .rpm; but, for now, Red Hat users will have to install things from a tarball. Since a fair number of the hits to this 'blog are from searches as to how to install Firefox 3.0 under RHEL 5.x, I'm going to infer that people are and will be surfing the WWWeb for instructions on how to install Firefox 3.5 under RHEL 5.x.

My first piece of advice is that one not install Firefox 3.5.1. When I tried using it, it would do something that caused the Linux user account to be logged-out. However, I've being trying version 3.5.2, and so far I've not had that problem with it. [Up-Date (2009:08/17): Unfortunately, I have since had some problems with version 3.5.2 logging me out of the system, and on one occasion it screw-up the display resolution. But these problems have not been so frequent as to move me to stop using this version.] That said, here are the steps that I recommend:

1. Download the archive, firefox-3.5.n.tar.bz2.
2. The tarball contains a directory, firefox, which should be dropped-in as a sub-directory of something. If you want to ponder where, then study the FHS. As for me, as root, I put it in /opt:

   tar -xjvf firefox-3.5.n.tar.bz2 -C /opt/

   (Replace that n with the actual number from the archive that you downloaded.)
3. Make sure that you have compat-libstdc++-33 (a Gnome C++ compatibility library):

   rpm -qa | grep compat-libstdc++-33

   If not, then as root install it:

   yum install compat-libstdc++-33

4. To avoid conflicts with SELinux, as root run

   chcon -t textrel_shlib_t /opt/firefox/libxul.so

   (If you didn't install the directory in /opt, or renamed the firefox directory, then you'll need to modify the above final argument to chcon accordingly.)
5. You'll need a .desktop file for Firefox (though you may already have one). As root, edit/create /usr/share/applications/firefox.desktop, ensuring that it reads

   [Desktop Entry]
   Categories=Application;Network;X-Red-Hat-Base;
   Type=Application
   Encoding=UTF-8
   Name=Firefox
   Comment='WWW browser'
   Exec='/opt/firefox/firefox'
   Icon='/opt/firefox/icons/mozicon128.png'
   Terminal=false

   (Again, if you didn't install in /opt, or changed the name of the firefox directory, then you'll need to change the above accordingly.)
6. Log out and back in or restart the system (to up-date the GUI).

I am amused by this eBay bid history:

I'll translate:

• On 27 July, at 11:30:10 PDT, seller posts item with an opening bid of $1.00 and some still unknown reserve price.
• At 19:36:05 PDT, first bidder enters a maximum bid of $6.16; this does not meet reserve price, so first bid is $1.00.
• On 29 Jul, at 12:45:05 PDT, second bidder enters a maximum bid of $5.00; entry automatically pushes bid of first bidder to $5.50.
• 10 seconds later, second bidder enters some maximum bid more than 50¢ than first bidder's maximum bid, and finds that his or her bid is now $6.66.
• 12 seconds later, second bidder enters some higher maximum bid, but his or her bid remains $6.66.
• Another 12 seconds later, second bidder enters some even higher maximum bid, but his or her bid remains $6.66.

If the second bidder were to enter a bid not less than the seller's reserve price, then his or her bid would become that. Otherwise, his or her bid will remain at $6.66 until some other bidder enters at least $7.16.

(BtW, I put the words reserve and maximum in quotes, because, as far as I'm concerned, eBay abuses each term, one way or another.)

The major muscles around my left shoulder — trapezius, deltoid, latissimus dorsi, and especially pectoral — have been hurting a great deal whenever I've moved that arm in the last few days. I hadn't recently engaged in any major physical activity nor been in an accident. I'm wondering whether I've injured a nerve with my back-pack.

---

On Thursday night or Friday morning, at the Hillcrest CVS/pharmacy, I noticed that bags of pistachios, regulary priced at US$3.99, were on sale for US$4.99 for those with a loyalty card. I brought this problematic sale to the attention of a supervisor, but the offer continued at least through Friday night.

---

On Friday night, I was walking to my car, when I spotted a feral mouse running ahead of me on the side-walk. All of the feral mice that I'd seen before were wood mice or deer mice, and all back in New Jersey, on property adjacent to woods. Here in San Diego, the only feral rodents that I'd seen were pack-rats (about the size of domesticated hamsters or gerbils[1]). This little creature looked like a house-mouse. At one point, the mouse was cornered in a door way. But, of course, I had no intention of hurting it and wouldn't even have wanted to catch it — it might have pups back in a nest somewhere, and I'd have to worry about the diseases that a wild mouse might carry.

---

This morning, near my home, I came across a relatively young immigrant man and woman, trying to figure-out how to start her car. They had another car with them. I asked if they needed jumper cables, and the man said yes, so I got mine. (Most people in this area don't have jumper cables; I keep long, heavy-duty cables in my car.) I let the guy do the connecting — I just discreetly watched to ensure that he didn't wire the batteries in series — because I neither wanted to make him seem ineffectual in front of the woman nor wanted to be blamed should something go awry.

He really didn't know what he was doing. He connected the negative line directly to the batteries, repeatedly clacked clamps together as a way of ensuring that the connection to the running car was good, didn't listen to the car that wasn't starting, and didn't seem to understand that a bad battery wouldn't explain the inability to start the parasitic car.

I couldn't hear the solenoid. I tried to explain to them that there was a problem with the ignition or a blown fuse. Anyway, they eventually gave-up on trying to jump-start the car, and returned my cables. I wished them good luck. The car is parked in a metered spot, but they have until 8 AM on Monday before that meter has to be fed.

---

I have an intermittent, vertical, purple line appearing on the display of my note-book computer. This tells me that the LCD panel, only about a year old, is beginning to fail. Bah!

---

[1] I'll rat myself out: One night several months ago, thinking that a rodent (which I could not see well) might be an escaped pet, I caught one of those pack-rats with my hat. A neighbor told me that the creature was a rat; it plainly wasn't a domesticated R. norvegicus, so I released it. The Woman of Interest noted to me that I needed to do something about mites and what-not that might have been transferred to the hat. I felt foolish.

If you’re actually trying to install another version of OpenOffice, then click on the OpenOffice tag, as there may be an entry on that other version.

My suggested procedure for installing OpenOffice 3.1.x under RHEL 5.x is essentially the same, mutatis mutandis, as that for installing OpenOffice 3.0.x:

1. If you don't have a JRE installed, then install one. Version 1.6.0 update 13 of the JRE was distributed with OpenOffice 3.1.0 when Sun was already at update 14. As I write, OpenOffice and Sun are in-synch at update 16, but check with Sun for a more recent version when you are installing OpenOffice. (I suggest that one use jdk-6uxx-linux-xxx-rpm.bin or jre-6uxx-linux-xxx-rpm.bin, rather than jre-6uxx-linux-xxx.bin.) The remainder of these instructions assume that one has a JRE installed.

2. Remove any earlier installation of OpenOffice. As root, enter these two commands:

   rpm -qa | grep openoffice | xargs rpm -e --nodeps
rpm -qa | grep ooobasis | xargs rpm -e --nodeps

3. Unpack OOo_3.1.x_LinuxIntel_install_wJRE_en-US.tar.gz (or the version appropriate to a devil-language, if you use one of those) to your filespace.

4. Go into resulting OOO31x_mxx_native_packed-x_en-US.xxxx/RPMS/ (or to the OOO31x_mxx_native_packed-x_xx-xx.xxxx/RPMS/ corresponding to your devil-tongue).

5. As root, run

   find . -maxdepth 1 -name "o*.rpm" | xargs rpm -U

6. As root, run

   rpm -U desktop-integration/openoffice.org*-redhat-menus-*.noarch.rpm

   (NB: You may need to log-out and back-in for the Applications menu to be up-dated and list the latest OpenOffice components. Your previous version may continue to be listed on the menu.)

7. As root, run

   rpm -U userland/*.rpm

8. Tell OpenOffice which JRE to use:

   • Launch OpenOffice:

     /usr/bin/openoffice.org3

     (It may not be listed on the applications menu unless you have logged-out and back-in. Before then, you may be able to launch it from the menu by way of a listing for a previous version.)
   • Select

     Tools | Options… | OpenOffice.org | Java | Use a Java runtime environment

   • Choose one of the environments that is then listed.
   • Click the OK button.
   • Shut-down OpenOffice. (The change will be in effect upon next launch.)

NB: This entry was editted on 2009:09/13, to make it compatible with OpenOffice version 3.1.1, and otherwise to improve the instructions. One of the improvements reflects the discussion in the first two comments to this entry.

The Windows partition on my computer has its own Linux .Trash folder for each account, which folder is not cleared when I select the Empty Trash option of the Trash icon of my Linux desktop. So I thought, Eh, I'll created some symbolic links in my home-folder, so that I can quickly access the /mnt/WindowsXP/.Trash-root and /mnt/WindowsXP/.Trash-daniel.

An amusing thing occurred when I tried to delete files from one of the .Trash folders with the GUI by opening the folder by way of the symbolic link, selecting the files to be deleted, and then using the Delete key.

The file is deleted, but it's replaced with a copy. Delete x.y, and it is replaced by x (copy).y; delete that and it's replaced by x (another copy).jpg, which would be replaced by x (3rd copy).y, and so forth.

(The same results obtain if one selects Move to Trash, but the result is a little less bizarre since one is then plainly telling the computer to move the file to a directory in which it is already listed.)

Convenient work-arounds are easy enough when I want to empty the whole .Trash folder by way of the GUI, but these symbolic links aren't nearly as useful as I'd hoped.