A SELECTION

Jerry,

You asked me to drop you a note about the Reformed English stuff I scribbled. Here you go. It's still in early stages.

DVD vs. DIVX: The Truth Is Out There => http://www.riva.com/dvd_divx.html

---------------------Read about "Reformed English"----------------------------

I don't know your national versus international readership, but on JONCON UK, German, Russian and Australian citizens have all said they wish to see an "electro-BYTE" (call it what you will).

The trouble with a cheque is the Bank charges me $6 to obtain a draft for US$10.

If you were able to take credit card, VISA or Mastercard have no fee, just a rather shaded exchange rate.

That's why I paid a three year subscription to BYTE last year - for all the risks, it reduced the ancillary costs.

Card - I'd do today. Cheque - oh, and add AU $1.50 for postage - I'll think about.

You get $10. I spend AU $16.4 + $6 + $1.50 = 23.90.

However, if you say on the site - no money, no entrance, I'll spend it.

But I'd really prefer VISA - AU $17.

regards

Following is an actual letter sent to Miller Brewing Company and their response:

Miller Brewing Company

Milwaukee, Wisconsin 53201

Dear Sir or Madam,

I have been a drinker of Miller beers for many years (actually, ever since the company that made my favorite beer back in the mid 80s donated a big chunk of change to Handgun Control Inc. so I had to switch). Initially, my beer of choice was Lite, but some time in mid-1990 while in Honduras I switched to MGD smuggled up from Panama. Now,for nearly six years, I have been a faithful drinker of MGD.

For these past years, I have come to expect certain things from Genuine Draft. I expect that whenever I see that gold can of MGD, I am about ready to enjoy a great, smooth brew. But wait! Sometime around the first of the year, my beloved MGD changed colors, so to speak. That familiar gold can was no longer gold! Knowing that I am, by nature, somewhat resistant to change, I forced myself to reserve judgment on the new can design. Gradually, I grew to appreciate the new label. That was until about May of this year. That was when I discovered (empirically) that I really didn't like the new design. Further investigation of the cause of my distress resulted in the following observations:

4. Sunlight striking the can causes radiant warming of the surface of the can. 5. The resultant heat (energy) is transferred through the aluminum, by conduction, to the contents of the can (the beer).

6. Warm beer sucks.

This is a process that can be observed in just about any beer. However, this process is significantly accelerated in MGD because you painted the damn can ... black!!!

Who was the rocket scientist that designed the new graphic for the can and implemented the change right before summer? Granted, this process may not be real evident up there in Wisconsin, but down here in Oklahoma (OR TEXAS) where the summers are both sunny and hot, this effect is quite a problem. There's no telling what the folks in Texas and Arizona are having to put up with.

Knowing that you would probably not address this issue unless you had firm evidence of a problem, I and several other subjects conducted extensive experimentation. The results of these experiments are listed below. The experiments were conducted over two days on the deck next to my pool. The study included seven different types of beer (leftovers from a party the previous weekend) that were initially chilled to 38 degrees F and then left exposed to sunlight for different lengths of time. These beers were sampled by the test subjects at different intervals. The subjects, all normally MGD drinkers, were asked at each sampling interval their impressions of the different beers. The length of time between the initial exposure to sunlight and the point where the subject determined the sample undrinkable (the Suck-point) was determined. This means that the higher the number at the suck-point, the longer it could sit in the sun without sucking. The average ambient temperature for the trials was 95 degrees F. Beer Type Color of can Average Suck-point(minutes)

It was evident that the color of the can directly correlates to the average suck point, except for Coors which was pretty much determined to suck at any point.

It is to be hoped that you will consider re-designing your MGD cans. All beer drinkers that are not smart enough to keep their beer in the shade will thank you.

The Miller response appears below. They have had a lot of fun with this guy's letter. Enjoy...

Dear Bradley Lee,

Thank you for your letter and your concern about the MGD can color as it relates to premature warming of the contents.

Like you, we at Miller Beer take beer drinking very seriously. To that end, we have taken your letter and subsequent experiment under serious consideration. Outlined below are our findings and solution to your problem. May we add that we have had similar letters from other loyal beer drinkers, mostly from the Southern United States.

First, let us congratulate you on your findings. Our analysis tends to agree with yours regarding Coors. It certainly does suck at about any temperature. Now, it was our intentions when redesigning the MGD can to create better brand identity and brand loyalty. Someone in marketing did some kind of research and determined we needed to redesign the can. You will be pleased to know, we have fired that idiot and he is now wreaking havoc at that pro-gun control beer manufacturer. The design staffer working in cahoots with the marketing idiot was also down-sized. However, once we realized this mistake, to undo it would have been even a bigger mistake. So, we took some other actions. From our market research, we found a difference between Northern beer drinkers and Southern beer drinkers. Beer drinkers in the South tend to drink slower than beer drinkers in the North. We are still researching why that is. Anyway, at Miller Beer, it was never our intentions to have someone take more than 2.5 minutes to enjoy one of our beers. We pride ourselves in creating fine, smooth, quick drinking beers and leave the making of sissy, slow sipping beers to that Sam guy in Boston. However, it is good to know that you feel our Miller Lite can last as long as 6 minutes. However, may we suggest in the future you try consuming at least two in that time frame.

>From your letter, we had our design staff work 'round the clock to come up with a solution that would help not just MGD but all our fine Miller products. We hope you have recently noticed our solution to your problem. We found that the hole in the top of the can was not big enough for quick consumption. So, we have now introduced the new "Wide Mouth" cans. We hope this will solve all your problems.

Might I also suggest that if you want to get the beer out of the can even faster, you can poke a hole on the side near the bottom, hold your finger over it, open the can, tip it to your mouth and then pull your finger off the hole. This is a common way to drink beer at parties and impress your friends. This technique is known as "shot-gunning". You should like the name.

Again, thank you for your letter and for bringing to our attention that there might be other beer drinkers taking more that 2.5 minutes to drink our beers. Let me assure you that I will have our advertising department work on campaign to solve this problem, too.

Jerry,

Just looked at the table test. ACK! :)

It's not very legible. Here are the problems with using tables for layout:

1. You force people to use the same size window that you use. Thus, if someone wants to use a smaller window on a smaller monitor, they have to scroll around a lot. For that reason, fixed-width tables are a Bad Thing.

2. When you try to print something that's been done with fixed-width tables, you usually find that the content creator has a printer with a wider carriage than yours, and stuff gets chopped unless you play with reductions. Another Bad Thing.

3. Especially on slower computers, tables take longer to render, making the site seem much slower. The test page was at least an order of magnitude slower to load on my Power Mac 7100/80AV using Netscape 3.04, as compared with your other pages.

4. Using a table for a button-bar is inelegant, because it's only useful if you keep the text on the page short-and, Jerry, we ALL know that you don't keep your text short! :) Once the page gets long, the buttons get scrolled offscreen. For that reason, if a button-bar is indicated, it's better to use frames. Besides, properly designed frames will reduce load times.

5. It's best to avoid changing text color where possible; some people are color-blind and odd colored text may be invisible to them. Besides, some people configure their browsers to ignore text color tags, which may make the page illegible if the background is too funky.

I find a common mistake of site designers is to think that HTML is supposed to be used the same way as PageMaker-to create a Layout. HTML is a content-description language, not a page-description language. If you use the HTML tags to mark up the features of your text, rather than the appearance, you'll get the best results. For example, the <B> bold tag is an appearance thing; the <STRONG> tag is better because it shows that you meant the text to have a strong emphasis, and it results in boldface on browsers that support bold, and some other form of strong emphasis on other browsers, such as Lynx. It also makes it easier to analyze pages automatically.

I've occasionally heard HTML compared to fonts on the early Macintosh:

people tried to use all the fonts at once, and wrote documents that looked like ransom notes. People who are just starting with HTML tend to do the same sort of thing. For your purpose, we're here for your text, not your layout... so keep it simple, which will keep it readable. If your readers want a different text color, or a different font, or a different column width, we can easily adjust our browsers... which was the intent of HTML.

Hope it helps...

p.s. In case you were wondering: my tool of choice for making Web pages is BBEdit Lite on my Mac. By keeping it simple, it's easier to write HTML by hand and get good results-and most of all, get the expected results. All the Web editors I've tried to weird things, because they try to apply the PageMaker layout paradigm to HTML, rather than applying the HTML paradigm to HTML.

Jerry,

Just looked at the table test. ACK! :)

It's not very legible. Here are the problems with using tables for layout:

1. You force people to use the same size window that you use. Thus, if someone wants to use a smaller window on a smaller monitor, they have to scroll around a lot. For that reason, fixed-width tables are a Bad Thing.

2. When you try to print something that's been done with fixed-width tables, you usually find that the content creator has a printer with a wider carriage than yours, and stuff gets chopped unless you play with reductions. Another Bad Thing.

3. Especially on slower computers, tables take longer to render, making the site seem much slower. The test page was at least an order of magnitude slower to load on my Power Mac 7100/80AV using Netscape 3.04, as compared with your other pages.

4. Using a table for a button-bar is inelegant, because it's only useful if you keep the text on the page short-and, Jerry, we ALL know that you don't keep your text short! :) Once the page gets long, the buttons get scrolled offscreen. For that reason, if a button-bar is indicated, it's better to use frames. Besides, properly designed frames will reduce load times.

5. It's best to avoid changing text color where possible; some people are color-blind and odd colored text may be invisible to them. Besides, some people configure their browsers to ignore text color tags, which may make the page illegible if the background is too funky.

I find a common mistake of site designers is to think that HTML is supposed to be used the same way as PageMaker-to create a Layout. HTML is a content-description language, not a page-description language. If you use the HTML tags to mark up the features of your text, rather than the appearance, you'll get the best results. For example, the <B> bold tag is an appearance thing; the <STRONG> tag is better because it shows that you meant the text to have a strong emphasis, and it results in boldface on browsers that support bold, and some other form of strong emphasis on other browsers, such as Lynx. It also makes it easier to analyze pages automatically.

I've occasionally heard HTML compared to fonts on the early Macintosh:

people tried to use all the fonts at once, and wrote documents that looked like ransom notes. People who are just starting with HTML tend to do the same sort of thing. For your purpose, we're here for your text, not your layout... so keep it simple, which will keep it readable. If your readers want a different text color, or a different font, or a different column width, we can easily adjust our browsers... which was the intent of HTML.

Hope it helps...

p.s. In case you were wondering: my tool of choice for making Web pages is BBEdit Lite on my Mac. By keeping it simple, it's easier to write HTML by hand and get good results-and most of all, get the expected results. All the Web editors I've tried to weird things, because they try to apply the PageMaker layout paradigm to HTML, rather than applying the HTML paradigm to HTML.

Alas Jerry! Sorry about the loss of your mail! I sent an e-mail yesterday in which I mentioned the exact same problem. Here's a quote:

"Last night I had problems inserting large amounts of text into a single cell. Word kept doing weird things to it. I suspect this is a problem with Word itself. My solution was to create more cells with the draw tool, then insert parts of the stuff that way. Another solution might be to insert into the HTML directly, but I'm not sure Word would then be able to interpret the document."

So, the error IS reproducible, not that that's much consolation. When I pasted a long document into a single cell, I could see it in Word, even after clicking save, but when I closed the doc and reopened (or looked at it with a browser) the whole cell was gone.

Perhaps someone still has the stuff in their cache and can "work offline," find it, and mail it to you? Fire up IE4, go into Offline mode (file menu) and see if you can find it in the cache, or got to c:\windows\Temporary Internet Files and see if it's in there. If you looked at it with IE4 once or twice, there might be a copy. (Same with netscape.) I searched mine, and didn't find one, for some reason.

Good luck. Wish I'd been able to warn you earlier! (Maybe Darnell has a backup?)

-K ()

In response to your email to Jerry Pournelle, as found at:

The AMD K6-3D was announced at the 1997 Microprocessor Forum, an industry conference run by the company I work for, MicroDesign Resources ( http://www.MDRonline.com/ ). The K6-3D was renamed the K6-2 when it was announced this spring.

Our newsletter, the Microprocessor Report, ran a detailed article on the K6-3D at the time, and we've run additional material on the 3DNow! instruction set that appeared in the K6-2.

There's a subtle point to be made here about paper vs. electronic publications. The Report is a paper publication, but our secret to prompt coverage is advance disclosure from the vendors. Vendors often set their release dates to coincide with our cover dates so we can have a detailed article on the street the same day a part is announced. This often means our analysis is published on paper even _before_ the electronic media get the raw material they'll use to write their stories. (And why our analysts often get quoted in those stories.)

I've also written about the K6-2 and 3DNow! for my column in Multimedia Systems Design magazine ( http://www.msdmag.com/ ), though MSD suffers from publication lag just like other paper magazines.

You might want to check out our Web site for information on the Microprocessor Report, since it sounds like information you could use.

Oh, and just to comment on a couple other points on that page in case Jerry wants to add this letter to it, I use and prefer Eudora for email; it's a great package, powerful and easy to use. Most importantly, it's no more complex than necessary to do the job. I'm a big fan of elegance in software design. I also use Dreamweaver as a Web-page editor for the same reasons. Other Web-page editors are as powerful, or more, but all the other decent editors produce really messy HTML code. Dreamweaver's output is clean and concise, and if you edit it manually, Dreamweaver won't mess with your changes. Jerry, I strongly recommend both packages.

. png

Sir,

soon to follow will be my snail mail with check, (or, god forbid, cash) in the amount of $10. I understand the limitations of your offer, and am unoffended as you have certainly made it clear this is a donation, NOT a subscription.

I look forward to your continued success, as I, and others like me, are the beneficiaries. Hopefully, your books and this web page will be able to continue without being mutually exclusive.

By the way, do you have any particular voice word processors to recommend? I have a friend with a serious wrist condition working on his doctoral thesis (linguistics, argh). He uses a Mac currently, but is willing to switch to Wintel if necessary. Thanks for any help you can offer.

Bryan Broyles

Are you rating any of your pages for handicap access?

Good Luck

Clark

Clark E. Myers

206-662-3399

M/S 19-HJ seat 2-122-2-2E3-B

"Never ascribe to malice that which is adequately explained by incompetence," said Napoleon Bonapart. There are also degrees of SNAFU short of incompetence which may suffice.

Jerry Pournelle

Hi Jerry

I think that having a chance to keep reading your columns would be worth the $10. As I live in Europe I don't know how to pay the $10 without paying the bank more than the original $10.

Any ideas?

Alex Keizer.

If I really have to, I'll wear the impost for the international money order, but as noted, it's not a reasonable method.

Jerry-

Re: "First, I cannot find any way to tell Outlook 98 that I want to change over to Internet Explorer as the default."

Did you try the "Internet" icon in the "Control Panel?" If you choose the Programs tab, it'll allow you to configure which programs get used, and there's also a checkbox for IE to check to see if it's the default. I imagine if you check that box, then start IE4, it'll ask and set itself as default.

I believe that Outlook '98 abides by this information.

I think the Control Panel>Internet icon is created when you install IE4, and is part of their system integration. It's the same dialogue as comes up when you choose, in IE4, View>Internet options....>

Good luck!

Keith Irwin

Ps. What makes it so tempting to go almost totally Microsoft is not so much the features of the individual components, but the way they work together. For instance, drag a Word doc from a window to the "Inbox" on the folder list in Outlook 98, or to the Calendar, or the Tasks item. Pretty interesting!

Jerry,

In the Control Panel, Internet Properties, Programs tab, you can tell IE to check to see whether it is the default browser. Click it on en start IE4. It will ask that if you want it as default. Works with IE3 so I think it will work with IE4.

Good luck.

Sent: Wednesday, June 24, 1998 5:11 AM

To: jerryp@jerrypournelle.com

Hereby a little program to convert Netscap bookmarks to IE favorites.

Good Luck

Wim

Hi,

first of all, keep up the got work - I was an avid reader of Byte (and your collumn, which usually made me buy new hardware...). I especially like reading about your struggling with Windows' quirks and instability (I'm a Unix fan who usually need to reboot his computer every 6 months or so to upgrade the hardware). Rest assured that I will send my $10 as soon as you find a way for foreign readers to do so.

Anyway, I'm looking for a way to upgrade my computer. It's a PPro-200, and I'd like to upgrade to a Dual PPro. My question is, is it possible to use two processors, each with a different amount of L2 cache ? (mine has 256Kb, but if I buy a new one it would be at least a 512 Kb one). No salesperson or tech support has been able to answer that question. Do you have any dual-PPro motherboard to recommend ? I would also need more than 4 SIMM slots, so I could recycle my current memory (4x32 Mb).

I subscribed to BYTE mostly to read Chaos Manor. Although I miss BYTE, I do have to say I like your new forum better. ;). Virtually realtime Pournelle! Yay!

You'd asked about converting your bookmarks over to IE4, I've attached a good utility for that. In case the attachment fails, here's the URL where this can be found.... http://hotfiles.zdnet.com/cgi-bin/texis/swlib/hotfiles/info.html?fcode=000H 62. Live Long and Prosper!

(See attached file: webfld.zip)

<<Now all I have to do is figure out how to convert those Netscape bookmarks to IE4>> Here's how to do it one by one:

You can also do it wholesale by downloading the Bookmarks Converter tool from Microsoft. The filename is winbm2fv.exe. Regards.

Dear Jerry,

I've followed your PC adventures in Byte for many years, going back to the days of the S-100 bus. I've generally been in agreement with your opinions as well. However, this time I felt a need to reply. Perhaps, its because the incident happened so close in time to my reading your column in the July 1998 issue of Byte.

I bought the Norton Utilities 3.0 and attempted to install and use it. In my system, it reported somerthing similar to you, perhaps 30 odd problems. I told NU30 to go ahead and fix them. It chugged for a while and proceeded to crash my system so badly that I had to re-install everything. Also, it somehow trashed my warm boot vector so that the system could never restart, it just ran in an endless BIOS initialization loop that reset itself with the AWARD PnP BIOS message.

My computer was running fine, I installed NU30 as a tool and it was like taking a sledgehammer to the instrument panel.

I think part of it is that this software now comes from Symantec. I have purchased several Symantec applications and found them to be extremely buggy and overly complex. Yes, I could access the Symantec web site and download the latest patches and fixes, but the original stuff that is put on on the CD-ROM should, at least, install properly. I spent three weeks with them, toll calls on the phone and such to get WinFax PRO 8.0 to install and run properly. What is surprising is that the old version of WFPRO for Windows 3.0 worked just fine.

Anyway, I am personally disappointed in NU3.0 and will not install it again. I loved the old DOS Norton Utilities and remember very fondly the hours I spend with Disk Doctor trying to recover files on my floppy disk or my 20 MByte, ST-506 drive. Obviously, things are orders of magnitude more complex now, but NU3.0, at least in my case, hasn't risen to the occasion.

I saw that another of your readers had experienced difficulties with Norton Utilities.

I installed Norton Anti-Virus, and it found a "changed" file, and offered to fix it. I allowed it to do so, and the computer then refused to reboot. In fact, the system now reported that the harddrive wasn't formated.

I called Norton. The answer, as I recall, was, sometimes a vfat boot partition is changed to a fat partition, resulting in your computer being unable to read the harddrive without a new formating. I asked how I could fix this. The techie replied, impatiently, you have to reformat. He said this happens, "sometimes". I expressed to him my displeasure with a protection software that would erase my harddrive, surely a result the program was designed to prevent. He reiterated that it happens sometimes, and acknowledged it wasn't mentioned in the manual.

I reformated, and began the laborious process of reinstalling from my backups. I lost 2 days of online bank transactions because at that time I backed up every 3 days. Now I back up my financial transactions after every transaction. I did NOT re-install Norton Anti-Virus.

Now, I use Norton's competitors, not because they are better, but because I got burned. I guess this is how competition thrives.

MAJ Bryan Broyles

After reading your last column where you praised NU3.0 thought that I ought to write and tell you that I didn't have such a nice experience.

Just read that I'm not alone just had to click on your mailto: to let you know.

It's the only software that I have ever returned. Once I got rid of it I was able to find the problems that it was reporting, but had done nothing to help me solve and had made it even harder to narrow down.

Jerry,

Your problems with the sluggish 486 sound familiar. I thought I'd offer my suggestions. By the way, I really enjoy the new online Chaos Manor and I look forward to you working out a payment method.

I've noticed that Norton Utilities for Windows 95 (I have experience with versions 2 and 3) can make a pretty big difference on a system with an aging hard drive. Run Speed Disk, but don't accept the defaults. Have it sort the files by last access date, grouped by week. Files that you access all the time move toward the front of the disk, where access is faster. Files that you never touch (Windows 95 has plenty of those) move to the slow part of the disk, where they belong.

Of course, initially running Speed Disk with these settings takes forever. So wait until you've had enough with writing for the day, then clean up the disk with Disk Doctor, run Speed Disk, turn off the monitor, and go on your way. The system will be faster in the morning.

If you've got Norton Utilites 3.0 or Nuts &; Bolts, use the registry optimizing tools while you're at it.

Then there's the old trick of telling Win95 it's a network server. Go into Control Panel --> System --> Performance --> File System and change "Typical role of this machine" to Network Server. If you're running Win95a, you'll need to patch the registry, since 95a sets the server settings incorrectly. Run Regedit, and scroll down to HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\Current Version\FSTemplates\Server\. NameCache needs to be A9 0A 00 00, while PathCache needs to be 40 00 00 00. Win95a has them reversed.

Of course, throwing newer hardware at the problem often helps. If you've got another old IDE drive kicking around somewhere that isn't doing anything, putting it in the system and putting Win95's swap file there (and maybe putting Word there as well) will boost performance.

None of these things will turn Old Cow into a Pentium, but they'll at least make Word load faster.

____________________________________________________________

Thanks for the kind words. I know I don't have to do it, and my wife points out I haven't been doing anything like as much fiction as I should. I'll have to be a LOT more selective in displaying and answering mail. Allocation of time, and all that. Darnell says the new data base driven system will be done Real Soon Now and then I won't have to work so hard at this. We'll see. Stay well, Jerry

Of course it will make you feel better to buy Starswarm! Roberta's web page can be found in my links; she has software that will teach people to read English. It works. Every time. At least every time we know of that it has been used, and that includes all ages, many intelligence levels, many first languages, and all socio-economic levels. If you can speak a thousand or so words of English you can learn to read it. Period. As to subscribing to BYTE, think of it as a new magazine. CMP has some good magazines, and perhaps they will be able to bring out a good new one with the BYTE name. But it won't be BYTE, which was unique, and damn it, I'm about to get worked up. I better go to lunch.

Incidentally, you caused me to take a look at King David's Spaceship at Amazon where I find it is out of print, so I have asked my agent if we can get the rights back. I know publishers who would love to keep it in print. I think I'd better check all my books. Thanks, and stay well.

Thursday, June 25, 1998

Yeah, and what a commentary on the program. There appears to be no way to force it to pay attention other than to hit on the head with a hammer. On the other hand, what it does it seems to do well. I have developed a routine in which I do all the editing in Word, use Commander to kill off all the backup files Word made (lest they be uploaded to the site too) then let Front Page publish. Seems to work.

There is a new item on using Norton Utilities and how to speed up old machines.

Well, thanks to you. I do seem to attract a lively and courteous group of associates. After a while there are enough nice smart people around that no one notices I remain an irascible SOB.

Anyway, I just thought that I would point out a couple of other pages that I visit regularly that have similar (not exact, but similar) goals. Ric Ford's Macintouch page is a really great resource for daily quick snippets of Macintosh related news. It is at http://www.macintouch.com/ Dave Winer also has a great site with a couple of different areas. The main page (http://www.scripting.com/) is just his links to either other writing that interests him or more of his detailed commentary. And then there is the DaveNet (http://www.scripting.com/davenet/ where he just has comments about whatever strikes his fancy. These DaveNet pieces used to be hosted at the HotWired site, and are updated on a varing schedule - sometimes more than once a day and at times the same one is up for more than a week.

dellis@halcyon.com Seattle, Washington

Thanks. I have found out lately there is a thing called "rings" in which people reference each others sites in a kind of chain. I doubt I'll ever to that myself, but there's a lot going on out on the web. More than anyone could keep up with.

private email: peterheberer@csi.com

I'm missing those too. As well as the time or staff to do them... I really am dancing as fast as I can. If we start getting money in here, I'll have someone trace links and insert them, Thanks for giving me the chance to explain why they aren't there already.

From: Peter N. Glaskowsky [png@ideaphile.com]
Sent: Thursday, June 25, 1998 8:32 PM
To: jerryp@jerrypournelle.com
Subject: Replaced email

Here's this email again. Third time's the charm, I hope.


Y'know, watching you work on your site almost in real time is like watching the Web itself get going in the first few years. :-)

Here's an incomplete answer to the "page or scroll?" question. Maybe it'll lead you in the right direction, however. You probably remember BBSs. I used to run one of the best in the country, the John Galt Line. All good BBSs had a feature that we've lost on the Web: the ability to present only the NEW material since the last time someone visited.

There are partial solutions, especially for message boards and similar BBS-like Web sites, but so far I haven't seen a good way to view only new Web pages on a normal Web site. It seems to me that this ought to be possible using cookies and server-side Java or Active Server Pages, but I'm not savvy enough to know how. This could be accomplished by changing a page to identify the new material, or by just not displaying the new material.

Short of presenting just the new stuff, you need to make sure everything is presented in a consistent format so readers can go through it quickly to find the new stuff, and offer a very clear way to signal where the new stuff is.

If you're going to modify a page from day to day, make sure you modify it only at the top, or only at the bottom, and use a visually distinct font to identify changes. Red isn't a distinct font, it's a distinct color. I'd recommend using something like this between each day's section:

<hr size="1" noshade>
<h3><font color="#CC0000">24 June 1998</font></h3> <p>

I'd also try harder to keep the text more consistent from one section to the next and one mail message to the next. I suppose you may just be experimenting to see what looks best, and that's okay, but I hope you plan to settle down eventually.

If you're looking for a consistent format for mail messages and your replies, I'd try putting the headers on successive lines, then a blank line, then the body the message on successive lines, with blank lines between paragraphs. The trick is to put <br> tags between lines and an extra <br> between paragraphs; this allows the font definition to remain intact instead of breaking at <p> tags. You could try pasting this into a page to see how it looks:

<font face="Courier New, Courier, mono" size="+1">
From: Reader Name<br>
Date: date<br>
Subject: Reader letter in monospaced font<br>
<br>
Body text text text<br>
<br>
Next paragraph end of text</font>
<p>
<font size="+1"><strong>Jerry replies:</strong></font>
<p>
Jerry's reply text in the default font<p>

I bet you can figure out some way to process your email into this format more or less automatically. BASIC would work.

Oh, one last thing. Your table work is getting better. I'd drop the third column of the table as used on your Space Mail page, since it doesn't really contribute anything and means the viewer's browser window must be four inches wider than the text, or so. Also, be careful when mixing lots of text in tables (not for the Microsoft cell-loss reason). Most browsers can start displaying plain text pages as they come down from the Internet. Some browsers need to see all the data that goes into a table before they know how to format it, causing them to appear to load much more slowly. For example, the Mail page takes about 15 seconds to load over my T-1 connection, but the first page of it shows up immediately so I can start reading. If it was in a table, it might take 15 seconds before anything showed up, a huge difference. You can usually fix this by putting fixed column widths into your table, but if you do, make sure the total isn't more than 630 pixels.

Oh, and something else I've noticed about your pages: you're using a lot of text which is smaller than the default Web font, which is already about as small as you can afford to use on a Web page. Size 3 is the default, but you seem to be using 2 a lot, and there are some things in size 1: "Adam and Tonya Engst" and "I have sent in my check" from your Mail page, for example.

You should NEVER use size 1 except in footnotes about automobile leases and the like. I would also never use size 2, but I concede that it may be occasionally useful.

. png

(This is the real end of the message. If you don't see this, let me know. :-)

Third attempt worked. I have the file, eliminated the old, put this in. Font sizes: I had to edit the html to get the tiny fonts out. Once Word and Front Page decide to start embedding tiny stuff in there, it don't go away. I did search and replace for size="1" to make it "2" then raised all the "2" sizes to "3". But 3 looks awfully big. Of course I have a big monitor.

Friday, June 26, 1998 A mixed bag that I won't be able to answer properly.

____________________________________________________________

______________________________________________________

Get Your Private, Free Email at http://www.hotmail.com

____________________________________________________________

From: Ward Gerlach [wgerlach@earthlink.net]
Sent: Friday, June 26, 1998 12:31 AM
To: Jerry Pournelle
Subject: Book Idea?

It's time and past time, and now what? I really am dancing as fast as I can. I used to have an editor to do a lot of this stuff. Oh well.

And now a bit of shameless self promotion...

I haven't been able to reproduce the error, but when I first went to the Microsoft site for Netscape users, I got script errors. I was first asked if I wanted to allow an Active X thing to run on my machine. I was then asked if I always wanted to trust Microsoft active x programs. I allowed this one, but I declined the affirmation of faith. Maybe Microsoft got mad at me for that? But the next time I went back to that site, it all worked.

Glad to:

There's a bit more here. And we have in fact been getting some money. Not a lot, but some.

Well, I can't agree. I think Tom Halfhill's technology articles were as fine as anything BYTE ever did. We may not have been all you wanted, but we were a lot better than anyone else. Ciarcia left over money issues. I would have kept him if it had been up to me, but he was asking for a lot more than they were paying me at the time and I didn't think I should quit over the issue. As to turning Circuit Cellar into a Rec Room, you'll have to ask Steve, since they had no more control over what he wrote than they did over me, which is to say, none.

Much of the rambling tech talk had to go because while readers loved it, ad agencies wouldn't buy ads. McGraw Hill insulates editors from advertisers, but publishers have to find ads or die. BYTE tried a lot of things in the hopes of gaining more paid space without losing our soul. As to timelessness, in this era is that possible? Things flow here so. Thanks, and stay well, Jerry

Thanks. Good point.

Precisely. Thanks.