Slashdot Mirror

I have quite a few friends and family who use computers, but are quite far removed from what's going on. They are probably only peripherally aware of Dmitry's plight, so I'm emailing them all this letter.

Also, I recently applied for a position as a software engineering manager at Adobe, which would be a good job for me and for which I feel I am qualified. Times have been tough for me and my little family and for quite some time I thought I might not speak out in a public way on this matter.

But long ago I decided that staying quiet was the wrong thing to do, so after quaking in fear for a while I decided I'd copy the following letter to the nice lady in the Adobe HR department who has been considering my application.

Subject: Free Dmitry

Friends,

I have long held the belief that computer programs are constitutionally protected free speech. They are, after all, how us programmers communicate with each other. This is also the opinion of at least one federal court, although it is yet to be tested by the Supreme Court.

However, on July 16, Russian computer programmer Dmitry Sklarov was arrested by the FBI for writing a computer program and presenting a paper on it at a security conference in Las Vegas.

His paper, "eBooks Security: Theory and Practice", exposed the woefully inadequate security schemes used to copy protect Adobe eBooks ("secure" electronic publications, basically encrypted PDF files).

If you have PowerPoint, you can get his presentation here:

http://www.download.ru/defcon.ppt

You can purchase, and download a free trial version of Advanced eBook Processor here:

http://www.elcomsoft.com/aebpr.html

Rather thank thanking him for revealing their engineering flaws, Adobe made a complaint to the FBI, and the FBI arrested him under the Digital Millenium Copyright Act. He is being held without bail, out of communication with his wife and children, in a foreign country, facing a $500,000 fine and five years in federal prison.

The digital millenium copyright act is clearly unconstitutional, not just in that it violates free speech for programmers, but that it violates fair use - the right of citizens to make limited copies of copyrighted materials for certain uses such as backup and academic research.

If you want to know more about Dmitry's case, please visit:

http://www.boycottadobe.com/

You'll find pictures there of Dmitry, and of his wife and children, who I am sure miss him greatly.

And please consider joining the Electronic Frontier Foundation, which is pressing two other court cases to try to have the DMCA ruled unconstitutional and will lend his support to Dmitry once the U.S. Marshalls tell them where he is, you can do so here:

http://www.eff.org/support/

Please pass this mail on to anyone who might be interested to hear it.

Ever Faithful,

Michael D. Crawford
GoingWare Inc. - Expert Software Development and Consulting
http://www.goingware.com/
crawford@goingware.com

Tilting at Windmills for a Better Tomorrow.

Mike

I have quite a few friends and family who use computers, but are quite far removed from what's going on. They are probably only peripherally aware of Dmitry's plight, so I'm emailing them all this letter.

Also, I recently applied for a position as a software engineering manager at Adobe, which would be a good job for me and for which I feel I am qualified. Times have been tough for me and my little family and for quite some time I thought I might not speak out in a public way on this matter.

But long ago I decided that staying quiet was the wrong thing to do, so after quaking in fear for a while I decided I'd copy the following letter to the nice lady in the Adobe HR department who has been considering my application.

Subject: Free Dmitry

Friends,

I have long held the belief that computer programs are constitutionally protected free speech. They are, after all, how us programmers communicate with each other. This is also the opinion of at least one federal court, although it is yet to be tested by the Supreme Court.

However, on July 16, Russian computer programmer Dmitry Sklarov was arrested by the FBI for writing a computer program and presenting a paper on it at a security conference in Las Vegas.

His paper, "eBooks Security: Theory and Practice", exposed the woefully inadequate security schemes used to copy protect Adobe eBooks ("secure" electronic publications, basically encrypted PDF files).

If you have PowerPoint, you can get his presentation here:

http://www.download.ru/defcon.ppt

You can purchase, and download a free trial version of Advanced eBook Processor here:

http://www.elcomsoft.com/aebpr.html

Rather thank thanking him for revealing their engineering flaws, Adobe made a complaint to the FBI, and the FBI arrested him under the Digital Millenium Copyright Act. He is being held without bail, out of communication with his wife and children, in a foreign country, facing a $500,000 fine and five years in federal prison.

The digital millenium copyright act is clearly unconstitutional, not just in that it violates free speech for programmers, but that it violates fair use - the right of citizens to make limited copies of copyrighted materials for certain uses such as backup and academic research.

If you want to know more about Dmitry's case, please visit:

http://www.boycottadobe.com/

You'll find pictures there of Dmitry, and of his wife and children, who I am sure miss him greatly.

And please consider joining the Electronic Frontier Foundation, which is pressing two other court cases to try to have the DMCA ruled unconstitutional and will lend his support to Dmitry once the U.S. Marshalls tell them where he is, you can do so here:

http://www.eff.org/support/

Please pass this mail on to anyone who might be interested to hear it.

Ever Faithful,

Michael D. Crawford
GoingWare Inc. - Expert Software Development and Consulting
http://www.goingware.com/
crawford@goingware.com

Tilting at Windmills for a Better Tomorrow.

Mike

I have quite a few friends and family who use computers, but are quite far removed from what's going on. They are probably only peripherally aware of Dmitry's plight, so I'm emailing them all this letter.

Also, I recently applied for a position as a software engineering manager at Adobe, which would be a good job for me and for which I feel I am qualified. Times have been tough for me and my little family and for quite some time I thought I might not speak out in a public way on this matter.

But long ago I decided that staying quiet was the wrong thing to do, so after quaking in fear for a while I decided I'd copy the following letter to the nice lady in the Adobe HR department who has been considering my application.

Subject: Free Dmitry

Friends,

I have long held the belief that computer programs are constitutionally protected free speech. They are, after all, how us programmers communicate with each other. This is also the opinion of at least one federal court, although it is yet to be tested by the Supreme Court.

However, on July 16, Russian computer programmer Dmitry Sklarov was arrested by the FBI for writing a computer program and presenting a paper on it at a security conference in Las Vegas.

His paper, "eBooks Security: Theory and Practice", exposed the woefully inadequate security schemes used to copy protect Adobe eBooks ("secure" electronic publications, basically encrypted PDF files).

If you have PowerPoint, you can get his presentation here:

http://www.download.ru/defcon.ppt

You can purchase, and download a free trial version of Advanced eBook Processor here:

http://www.elcomsoft.com/aebpr.html

Rather thank thanking him for revealing their engineering flaws, Adobe made a complaint to the FBI, and the FBI arrested him under the Digital Millenium Copyright Act. He is being held without bail, out of communication with his wife and children, in a foreign country, facing a $500,000 fine and five years in federal prison.

The digital millenium copyright act is clearly unconstitutional, not just in that it violates free speech for programmers, but that it violates fair use - the right of citizens to make limited copies of copyrighted materials for certain uses such as backup and academic research.

If you want to know more about Dmitry's case, please visit:

http://www.boycottadobe.com/

You'll find pictures there of Dmitry, and of his wife and children, who I am sure miss him greatly.

And please consider joining the Electronic Frontier Foundation, which is pressing two other court cases to try to have the DMCA ruled unconstitutional and will lend his support to Dmitry once the U.S. Marshalls tell them where he is, you can do so here:

http://www.eff.org/support/

Please pass this mail on to anyone who might be interested to hear it.

Ever Faithful,

Michael D. Crawford
GoingWare Inc. - Expert Software Development and Consulting
http://www.goingware.com/
crawford@goingware.com

Tilting at Windmills for a Better Tomorrow.

Mike

I think it's important that some respected member of the open source or free software communities should take the trouble to raise the issue of antitrust violations in this new license to the state attorneys general, as well as the U.S. attorney general - all of the parties in the U.S.DOJ et al vs. Microsoft suit.

Ranting about how unfair it is on slashdot will just create a lot of hot air and goatsex links. Getting Bruce Perens, Eric Raymond or Linus Torvalds to write letters addressed to each of the attorneys general could have Microsoft's copyright in their SDK taken from them for abuse of copyright, as well as having extra sanctions pressed against it.

Mike

A friend in another country wants to try Linux. I pointed him to where he could download the ISO's for SlackWare and Mandrake (shipping would take a long time for him, but he has a good net connection).

He doesn't have an existing Linux installation yet so he can't use cdrecord to make his bootable ISO9660 CD's. Are either Easy CD Creator or Nero for Windows capable of burning the bootable Linux iso images correctly?

The reason I'm concerned is that when Win98 scragged my laptop I got the ISO for tomsrtbt and tried to burn it with Toast on a Macintosh, and it wouldn't boot. Toast didn't seem to understand the ISO format tomsrtbt's image used.

Mike

I've been using SlackWare continuously for years. Unlike many people who started out with it and moved on, I started with Yggdrasil and moved to slackware at the recommendation of a friend, using it since 3.2 or so on a 33 Mhz 386.

Now I use it on my laptop and on my desktop PC.

I tried out Debian PPC on my Macintosh, and while it was nice to be able to run an install over the Internet, Debian was much harder to install than SlackWare. One thing that turned me off of debian was that after installing the full X11 package, debian enabled XDM - but this was before X actually was gotten to work on my system (the ADB mouse was frozen) so doing an apt-get made it impossible for me to actually log in to my computer. It really sucked undoing the damage that Debian's X11 maintainer did to my PPC system.

If you're running Slack 7.1 or earlier, a real good reason to update is that 7.1 comes with GCC 2.91.66. This version of GCC has a bug in the C++ parser that makes it emit an error for a certain piece of legitimate C++ code a friend wrote. I was able to FTP several packages off of ftp.slackware.com out of SlackWare-current and get gcc 2.95.3, which fixed the bug.

As for Slack not having package management, I don't know what you're talking about, I use pkgtool, installpkg and removepkg all the time to manage packages on my systems. If you have a bug, you can always try getting the pre-release packages for later versions of the software from slackware-current.

One nice thing you can do is download the source to a package off of the Slackware FTP site (or get it off the source CD), then get a later version of the source from the project's home FTP or CVS server, and use the scripts provided by Patrick to make a slackware package of some hot-off-the-presses code.

One project I have in mind is to build my own Slackware distribution, using the sources provided on a CD, except that I'd select Pentium III optimization in all the GCC command lines. If you do this with everything but the static libs that go into the distribution library directories, your system should run faster.

Note that I'm not a complete Slackware zealot. I can run through a Slackware install or upgrade pretty quickly, but it took me a while to figure out how to do it right. Recently when a colleague said he wanted to try Linux, I recommended he use Mandrake, but encouraged him to use Slackware for production systems.

Mike

I've been using SlackWare continuously for years. Unlike many people who started out with it and moved on, I started with Yggdrasil and moved to slackware at the recommendation of a friend, using it since 3.2 or so on a 33 Mhz 386.

Now I use it on my laptop and on my desktop PC.

I tried out Debian PPC on my Macintosh, and while it was nice to be able to run an install over the Internet, Debian was much harder to install than SlackWare. One thing that turned me off of debian was that after installing the full X11 package, debian enabled XDM - but this was before X actually was gotten to work on my system (the ADB mouse was frozen) so doing an apt-get made it impossible for me to actually log in to my computer. It really sucked undoing the damage that Debian's X11 maintainer did to my PPC system.

If you're running Slack 7.1 or earlier, a real good reason to update is that 7.1 comes with GCC 2.91.66. This version of GCC has a bug in the C++ parser that makes it emit an error for a certain piece of legitimate C++ code a friend wrote. I was able to FTP several packages off of ftp.slackware.com out of SlackWare-current and get gcc 2.95.3, which fixed the bug.

As for Slack not having package management, I don't know what you're talking about, I use pkgtool, installpkg and removepkg all the time to manage packages on my systems. If you have a bug, you can always try getting the pre-release packages for later versions of the software from slackware-current.

One nice thing you can do is download the source to a package off of the Slackware FTP site (or get it off the source CD), then get a later version of the source from the project's home FTP or CVS server, and use the scripts provided by Patrick to make a slackware package of some hot-off-the-presses code.

One project I have in mind is to build my own Slackware distribution, using the sources provided on a CD, except that I'd select Pentium III optimization in all the GCC command lines. If you do this with everything but the static libs that go into the distribution library directories, your system should run faster.

Note that I'm not a complete Slackware zealot. I can run through a Slackware install or upgrade pretty quickly, but it took me a while to figure out how to do it right. Recently when a colleague said he wanted to try Linux, I recommended he use Mandrake, but encouraged him to use Slackware for production systems.

Mike

I've been using SlackWare continuously for years. Unlike many people who started out with it and moved on, I started with Yggdrasil and moved to slackware at the recommendation of a friend, using it since 3.2 or so on a 33 Mhz 386.

Now I use it on my laptop and on my desktop PC.

I tried out Debian PPC on my Macintosh, and while it was nice to be able to run an install over the Internet, Debian was much harder to install than SlackWare. One thing that turned me off of debian was that after installing the full X11 package, debian enabled XDM - but this was before X actually was gotten to work on my system (the ADB mouse was frozen) so doing an apt-get made it impossible for me to actually log in to my computer. It really sucked undoing the damage that Debian's X11 maintainer did to my PPC system.

If you're running Slack 7.1 or earlier, a real good reason to update is that 7.1 comes with GCC 2.91.66. This version of GCC has a bug in the C++ parser that makes it emit an error for a certain piece of legitimate C++ code a friend wrote. I was able to FTP several packages off of ftp.slackware.com out of SlackWare-current and get gcc 2.95.3, which fixed the bug.

As for Slack not having package management, I don't know what you're talking about, I use pkgtool, installpkg and removepkg all the time to manage packages on my systems. If you have a bug, you can always try getting the pre-release packages for later versions of the software from slackware-current.

One nice thing you can do is download the source to a package off of the Slackware FTP site (or get it off the source CD), then get a later version of the source from the project's home FTP or CVS server, and use the scripts provided by Patrick to make a slackware package of some hot-off-the-presses code.

One project I have in mind is to build my own Slackware distribution, using the sources provided on a CD, except that I'd select Pentium III optimization in all the GCC command lines. If you do this with everything but the static libs that go into the distribution library directories, your system should run faster.

Note that I'm not a complete Slackware zealot. I can run through a Slackware install or upgrade pretty quickly, but it took me a while to figure out how to do it right. Recently when a colleague said he wanted to try Linux, I recommended he use Mandrake, but encouraged him to use Slackware for production systems.

Mike

I have some job hunting tips at:

• The Santa Cruz County Computer Industry Index
• Market Yourself - Tips for High-Tech Consultants

Best of luck to everyone. I have a lot of respect for a company that would use a program like Cerberus to stress-test its systems.

where else can you buy hardware that you can count on not being crappy?

Mike

I have some job hunting tips at:

• The Santa Cruz County Computer Industry Index
• Market Yourself - Tips for High-Tech Consultants

Best of luck to everyone. I have a lot of respect for a company that would use a program like Cerberus to stress-test its systems.

where else can you buy hardware that you can count on not being crappy?

Mike

I have some job hunting tips at:

• The Santa Cruz County Computer Industry Index
• Market Yourself - Tips for High-Tech Consultants

Best of luck to everyone. I have a lot of respect for a company that would use a program like Cerberus to stress-test its systems.

where else can you buy hardware that you can count on not being crappy?

Mike

I had a stove in the house I used to live in in Idaho when I was a kid that would pick up some AM station.

And later I lived a few houses down from someone with a whopping big ass (and illegal) CB antenna. I could pick him up clearly with a 2-inch audio speaker with a foot of copper wire hanging off of each of the terminals.

What was really a drag was when I'd try to tape a record album on my stereo, and the dickhead would decide it was time to chat with his buddies. My tape would record his transmissions.

The neighborhood got up a petition to the FCC to disclipline him, but nothing ever happened.

Someone told me that the stove couldn't possibly have worked because there was nothing to rectify the signal, but there it was doing it. Creepy too. It wasn't very loud but you could make out words in the broadcasts.

Mike

If you want to do some multithreaded programming, a nice way to do it in C++ is to use ZooLib. Here the threads and locks are easy-to-use C++ objects.

ZooLib supports Linux, BeOS for x86, Mac OS PowerPC and 68k, and Windows out of the box. It can be bound to other platforms in a straightforward way.

I believe the Mozilla framework is multithreaded as well.

Mike

The Neilson Norman Group did a study of real users' experience with wap phones a while back.

Reading the summary, and having a lot of respect for what the authors had to say on other topics convinced me it wasn't worth my while to bother with WAP.

The WAP Usability Report (available in PDF form for $26) reports on a study where 20 people were given WAP enabled phones for a week and asked to report back on their experiences. The study was done in London because of the advanced state of WAP services there. Read the summary here.

• 70% of users reported they would not use WAP within a year
• One user calculated it was cheaper to buy a newspaper and throw away everything but the TV listings rather than use WAP to check the BBC schedule

The summary says:

our basic conclusion is that WAP usability fails miserably; accomplishing even the simplest of tasks takes much too long to provide any user satisfaction. It simply should not take two minutes to find the current weather forecast or what will be showing on BBC1 at 8 p.m.

These are the same guys who test out concepts in web page design by sitting real users in front of browsers and watching them use the net. You may be familiar with some of the principles:

• Jakob Nielson
• Don Norman, author of The Design of Everyday Things, Things that Make Us Smart and The Invisible Computer
• Bruce "Tog" Tognazzini, one of the original designers of the Macintosh UI and author of Tog on Interface. Check out Tog's Software Design Bookstore to learn how to write software that doesn't suck. Read Top 10 Reasons the Apple Dock Sucks.

I strongly recommend anyone producing either content or software for the web regularly read Jakob Nielson's http://www.useit.com, where you could have found out before you invested that neither WAP nor advertising on the Web work.

I link these and a couple other useful sites in my brief section on Some Web Application Design Basics in Use Validators and Load Generators to Test Your Web Applications:

I'm not talking about pretty rollover buttons here, folks.

You need to understand that many web sites are developed with investments totalling many millions of dollars, only to have the effect of driving away any user who might have the misfortune to stumble across them, with much resulting heartbreak and the loss of fortunes.

Mike

It's not as stupid as you think, the first telescopes grew out of eyeglass making.

While you could make a telescope this way, you wouldn't get one with a very wide objective lens, so it wouldn't capture much light.

Also, you have the problem that the lens is made out of only one kind of material (plastic these days), so you have the problem of chromatic aberration - any one kind of lens material refracts light of different wavelengths differently, so you can't get all the colors in focus simultaneously and you get colored fringes around your stars, or just blurring of larger objects.

Mirrors work by reflection, and reflect geometrically, so they don't suffer from chromatic aberration. I think this is what Newton was after when he invented the newtonian telescope.

A lens suitable for astronomy needs at least two components, and needs to be made of glass of precise characteristics, and free of internal defects, and their are four surfaces to grind and polish, and further you can only support them at the edges. So it's impractical to make large quality refractor lenses - the biggest is the Yerkes observatory 40 inch.

Modern observatory instruments are all mirrors, I think the biggest single mirror is the 236 inch in Russia, and you can also compose a mirror out of numerous smaller mirrors, and the biggest one of those is the Keck 10 meter on Mauna Kea, Hawaii which has thirty mirrors in it.

Mike

On the other hand, I wonder what it sounds like at singles bars when you say to women "you know, I can grind my own telescope mirror". :)

Uh, yeah, many take it as evidence of unabashed Geekness.

While there are some women who grind telescope mirrors, it still seems to be a largely male activity, and while my wife encourages it, she thinks it's a pretty odd thing to get excited about.

I read a lot on the ATM list about how wives tend to think their ATM husbands are pretty weird for making telescopes.

On the other hand, I am married, so it can't be all that unattractive to the opposite sex. Probably best something not to put forward on the first date though.

Mike

Here are the basics of how you grind a telescope mirror. There are many variations. You definitely want to get a book. I used:

• Amateur Telescope Making by Albert Ingalls
• How to Make a Telescope by Jean Texerau

and one other I can't find anymore. There are other good books.

The stubby Celestron and Meade telescopes that are popular with amateur astronomers who prefer to purchase their instruments are of a type called a "schmidt-cassegrain". This has a nearly flat corrector plate in the front, that actually has a shallow fourth-order curve ground into it to correct spherical aberration, a deep prolate spheroidal primary mirror, and a convex secondary mirror mounted on the back of the center of the corrector plate.

It's the convex secondary that makes the telescope a cassegrain. The 200 inch on Palomar is a cassegrain. I don't have a schmidt-cassegrain to show you but here's how an ordinary cassegrain is laid out.

The use of the schmidt corrector plate allows one to make the telescope very short, with a small ratio of focal length of the primary to its diameter, without making images away from the center of the field blurry.

This is an advanced kind of design for an amateur to make oneself, although many amateurs have. Here's how one guy made a schmidt corrector plate.

The typical amateur starter scope is the "newtonian reflector". This has a concave parabolic mirror at the back end of the tube, and a few inches inside its focus is an optically flat mirror at 45 degrees. The optical path shown in the diagram is for the light from a single star, an image is formed from light sources spread across a small angle, and a small image is formed at the focal plane where it's examined by the eyepiece (a high-power magnifier) or photographed with film, a CCD or I guess even a webcam.

If you make a parabolic mirror with too short a ratio of focal length to diameter (the f-number, like the f ratio on a camera lens), then the images away from the center are blurred. This is called "coma". A parabola only focuses light perfectly if it's parallel to its axis and tilting the beam introduces coma. A ratio of 1 to 4 is about the shortest you can make it - f/4. My 6 inch is f/8, my 10 inch is f/3.5, and my 8 inch is f/6.

Having a longer focal length gives you greater magnification. Having a shorter one gives you a wider field of view, within the limits of the coma. Having a shorter focal ratio also makes it easier to fit in a car, an important consideration for making the scope enjoyable. Those Celestrons are nice because they'll easy fit in the trunk of a car or even in airline luggage (with a hard case) but it comes at the expense of a fancier design.

For the first homebuilt scope one usually grinds the primary and buys the flat diagonal mirror from a vendor. More advanced amateurs make their flats too but again optically flat surfaces are hard to make.

Making a primary that doesn't have too short a focal ratio is not too bad because the grinding process naturally makes a sphere. You grind a sphere of the right radius of curvature, fine grind through successively finer grits, then polish. You then use an optical test to get the mirror perfectly spherical, then deepen the center to move from a sphere to a parabola of revolution, testing carefully as you go.

The way I ground my mirrors was with pyrex mirror blanks on plate glass tools. Initially each is flat. They are both pretty thick, my 8 inch is about 1.25 inches thick, to stiffen them so they don't lose their figure. You have to have a figure that is perfect to about 1/8 of a wavelength of visible light in variation across the whole face of the glass, so any bending is disastrous. The 1/8 wave limit is the same for mirrors of all sizes so it's much harder to figure larger ones - best to start small. I would recommend an 8 inch for a first mirror. I have heard of people doing much larger first telescopes though.

What you do is sprinkle some granulated silicon carbide and water on the tool, place the mirror blank face down on it and push it back and forth until the silicon carbide ("carborundum") breaks down. (This is the same abrasive as you find on black wet-or-dry sandpaper, only in free-flowing powdered form). Then you add more abrasive and water and repeat. When too much mud builds up you wash it off and add more abrasive again.

To grind a concave curve into the mirror blank you place it on top, face down, grind with long strokes and have it hanging mostly off the side. Also you put pressure on it, either pushing hard or putting weights on it. This concentrates the grinding action in the middle and a shallow sphere develops.

Every few strokes you rotate the mirror a little, and once a minute or so you rotate the tool a little, with the idea that every part of the mirror gets ground over every part of the tool in every direction.

These days it has become popular to "hog out" a mirror with a metal ring tool, like a pipe cap as I'm about to try, then after rough grinding you make a fine-grinding tool out of small bathroom tiles mounted in dental stone or portland cement. This is in part because it's getting harder to get telescope making kits, unfortunately because it's so easy to buy a Celestron people don't make their own as much anymore. So people conserve the glass just for the mirrors and make the tile tools instead.

Be aware, before you say "well it's easier to buy a Celestron", that the price of a telescope goes up astronomically with increasing diameter - my 8 inch kit was $78 including shipping, I'll probably spend a few hundred to make a nice clock-driven mount, but the 10 meter telescope on Mauna Kea cost $90 million! If you know how to make your own, it is within your reach to grind your own 20 inch, which will have astounding views, but few of us could hope to afford to purchase a twenty inch commercial scope.

I know people who have ground 30 inch scopes and I know of some amateurs who are now figuring a 67 inch mirror!

Anyway it takes several hours of work to rough grind your mirror, more if you're doing a short f/number, less if you have a higher one, also less for smaller mirrors and more for larger ones. My 6" f/8 was about as deep as the thickness of an american dime, I don't know a little more than a millimeter.

Then you fine grind, grinding for a few hours with successively finer grades of abrasive. Usually you rough grind with 80 mesh silicon carbide - it is graded by sieving it through a mesh with 80 wires in it (same as the sandpaper sizes). Then you grind with #120, #220, #320, #400 and then several very fine grades of aluminum oxide whose sizes are given in microns.

The idea is that each finer grade erases the pits left by the previous grade. Between each grade you must scrupulously clean yourself, the mirror and tool and your work environment lest a coarse particle get into a finer stage and cause a scratch.

With each grade the mirror and tool surfaces will become more and more accurately spheres, within the limits of the sizes of the grits. This is because a sphere is the only shape that allows two surfaces to be placed anywhere against each other in any position or rotation (a flat surface is the limit of this as the radius goes to infinity). If there are any high spots, they will get more pressure and grind off quickly; any low spots will miss out on grinding and the surrounding surface will come down to match.

Then you polish. You make a "pitch lap", using either another dental stone base or the glass grinding tool, covered with refined, thickened pine pitch. You cut channels in the pitch with a knife or mold them in with a silicone mold. Then you cover the pitch lap with a suspension of cerium oxide in water, or else ferrous oxide (same as rust but finely powdered - "jeweler's rouge"). Then again you stroke the mirror on the pitch lap.

During fine grinding and polishing you use shorter strokes, and alternate which is on top, the mirror or the tool, to keep the depth constant. You also stroke a little side-to-side, in a W pattern. This evens everything out.

To test the mirror you use the Foucault test or the Ronchi Test. The foucault test appatatus I link to is much fancier than you need, although nicer to use - you can do it all with your naked eye and the tester, you don't need a camera.

In each test you use a light emanating from a pinhole or narror slit just to the side of the center of curvature of the mirror. The image of the pinhole or slit will form an equal distance to the other side, where you can place a knife edge (Foucault) or screen (Ronchi) across it and hold your eye there and look at the mirror.

It's kind of hard to explain but each of these has the effect of dramatically magnifying deviations from spherical surfaces in the mirror. A dramatic demonstration is to have someone hold their hand in the beam - you can see the distortion in the beam caused by the warm air rising from their hand.

You can easily make out a bump or hollow that's a fraction of a wavelength high on the glass.

Then you make your mirror perfectly spherical by preferentially polishing off the high spots. If you did the fine grinding and polishing well you won't have to work hard to do this.

Unfortunately what we want is a parabola, not a sphere. This must have a precisely controlled error in each test. This is a little more than I want to get into, but basically your preferentially polish out the center of the mirror so it's deeper in the middle than appropriate for a sphere by a little bit. Get it just right and you have a parabola, and your mirror will focus perfectly.

Then you package it securely and send it off to one of the people who does vacuum aluminization. They clean the mirror extremely well, place it in a high vacuum, and evaporate aluminum off of tungsten wires. The aluminum vapor sticks to your glass and you have a telescope mirror.

Mike

When I was in junior high and high school I ground, polished and figured several telescope mirrors. I did a 6 inch, then a 10 inch, and finally an 8 inch.

The 6 inch had a decent figure but I didn't know I could send it away to be vacuum aluminized, so I chemically deposited silver on it using chemicals I bought at the University of Idaho chemistry stockroom. Take my advice, it's much better to get a mirror aluminized.

I hurried a bit too much on fine grinding the 10 inch and wasn't happy with it, so I tried again with my 8 inch and was much more patient, and got excellent results from it (1/10 wave according to Chabot Amateur Telescope Maker's Workshop's Paul Zurakowski).

Grinding telescopes and being a sciency kind of guy led me to study astronomy at CalTech where I assisted CalTech astronomer Jeremy Mould in observing the the Palomar 60 inch and 200 inch telescopes - the experience of a lifetime for an amateur astronomer.

It's been about 18 years since I last worked any glass but I just bought an 8 inch plate glass kit from Dan Cassaro. You can buy Pyrex kits and optical glass (suitable for lenses) from Newport Glass.

I'm starting to write about the telescope I'm about to work on here.

If you are in the San Francisco Bay Area check out the Eastbay Astronomical Society's Chabot Amateur Telescope Maker's Workshop (there's an observatory there too, it's in Oakland), Fremont Peak Observatory, which has a 30 inch reflector that's open to the public, with regular gatherings of amateurs who bring their telescopes up there, and the San Francisco Sidewalk Astronomers - the Sidewalk Astronomers set up telescopes on city sidewalks and introduce people to astronomy by inviting them to look through their scopes.

You can get books on astronomy, and importantly, the specifics of how to actually grind and polish a telescope from Willman-Bell and Newport Glass.

Check out this guy who made a ribbed mirror blank by cutting out a pattern from one disk of glass with a water jet and fusing it to a solid sheet in a furnace.

Visit Google's index of Amateur Telescope Making, particularly http://www.atmpage.com.

If you want to get into amateur telescope making, take advantage of an immensely valuable resource that wasn't available to me when I was a kid - subscribe to the ATM List - here's the FAQ.

Mike

At Apple most people could say what they pleased on their cards, with the manager's approval.

I held a debugging job, and picked "Cybernetic Entomologist".

Mike

The Yellow Network Coalition takes donations of old computers, mostly 486's, and refurbishes and tests them, installs linux, and donates them for use as IP masquerading servers and user workstations.

I gave them my 486/100 after many years of hard service, and they have it in use somewhere.

They're good people, and give free lessons on how to do network administration, firewalling, linux administration and the like.

Mike

From a quick look, some of the problem is the handling of certificates, and an HTML validator won't help that, but it will help a lot of other problems with standards compliance that make your page not show up in one of the many browsers that are available.

Using a validator during daily development of your website, whether static or dynamically generated pages, in the long run just makes it a lot easier because you catch a lot of careless errors. Imagine how hard it would be to write syntactically correct C code without using a compiler!

• Validate Your HTML here
• Validate your CSS here
• Use Validators and Load Generators to Test Your Web Applications

If you'd like to try validating a couple pages real quick then click here to validate:

• http://slashdot.org
• http://www.debian.org/

Mike

SPAM is a registered trademark of Hormel Foods and should be kept off the Usenet News..

I notice now that my link to Hormel Foods Corporation vs. Jim Henson Productions Inc. (the opinion in a lawsuit) is now dead. Anyone got a good link?

Mike

You can find a list of application frameworks, many of which are cross-platform and many of which are open source, at the GUI Toolkit, Framework Page.

Their forwarding link at http://www.theoffice.net/guitool seems to be down but the original at Geocities is still up.

Please also read my essay on why it is important to write cross-platform code - with quotes from Judge Jackson on why Microsoft felt it was important enough to put a stop to cross-platform development that it broke the law.

My favorite cross-platform application framework is ZooLib, written by my friend Andy Green and his clients Learning in Motion. It allows you to write a single C++ sourcebase and deliver multithreaded native executables for Mac OS, BeOS, Windows and Linux/XWindows.

Mike

For details on the 128 Gig limit, the 1024 cylinder limit and other horrors from our DOS legacy, see the Large-Disk-HOWTO, particularly History of BIOS and IDE Limits.

Mike

If you can resolve a host, it may be that the host IP is in your nameserver's DNS cache.

I can resolve www.hotmail.com (it redirects me to somewhere at passport.com) but I find that I cannot ping either of hotmail's nameservers. I think hotmail is in my ISP's DNS cache.

I can ping all four of apple.com's nameservers, and I can also resolve www.apple.com. Can you ping any of apple's nameservers? See their Whois entry.

I cannot ping either of gnu.org's nameservers, and I cannot resolve any host in the gnu.org or fsf.org domains. See my post below about this.

Mike

Perhaps you'd like to consider submitting a claim to The Commercial Lawyer's Network, a commercial collection agency that has attorneys on staff.

I got burned for a few weeks pay by a consulting client last year and they never answered my attorney's attempts to collect so I referred them to ComLawNet.

Try their handy online claim submission form. You will need to agree to their agreement.

They seem to have a certain... familiarity... with dot-coms.

Read their FAQ. Note that they emphasize your chances of collecting are much greater the earlier the debt is placed into collections.

I'm still waiting for my money but I only placed them into collections a couple weeks ago.

Mike

If you want to know more about filesystems, a very good book is Practical Filesystem Design with the Be Filesystem by Dominic Giampaolo, ISBN: 1558604979

Dominic is the designer and implementor of the BeOS BFS filesystem, a multithreaded journaled 64-bit filesystem that supports indexed file attributes.

In the beginning of the book there is some good discussion comparing and contrasting various filesystems, including what is publicly known about NTFS. Other filesystems discussed include BSD's FFS, Linux ext2, Mac OS HFS and Silicon Graphics XFS. He also discusses some of the basics of designing any filesystem, the general approaches taken towards filesystem design, and discusses the BFS in depth. It discusses much more modern and advanced topics in filesystem design than are covered in most operating system texts, like journaling and accessing the filesystem from a multithreaded kernel.

By "attributes" I mean extra chunks of data that live outside the main data sequence, and are used for such things as denoting the MIME-type of the file in the filesystem. By "indexed" I mean that an application can tell the OS that it wants indices created for particular attributes, and then applications can do boolean queries on the attributes and get the responses quickly.

There is a linux version of the BFS filesystem available as a patch - it is not yet in the main kernel tree, and I believe it is read-only. It is complicated to make it writeable because it is a journaled filesystem. You can get it here.

To try out the Be filesystem fully, get the free-as-in beer version of the BeOS from http://free.be.com/ to install under Windows or ftp://ftp.be.com/pub/beos/ to get the version you can install under Linux. The BeOS personal edition creates a virtual filesystem within a regular file on FAT or ext2.

If you have a partition to spare, much recommended is BeOS 5 Pro which you can inexpensively purchase from Gobe Software. The Pro version can also do symmetric multiprocessing.

Mike

Mike

If you really care about the GPL license on your software, then it is very important that you assign the copyright to the Free Software Foundation..

That is, unless you have a lot of money and enjoy fighting lawsuits.

That's because only the copyright holder of a work whose copyright has been violated has standing to sue. While I might imagine that the FSF might be willing to lend you some legal advice if you hold the copyright to your work, they don't have standing to sue themselves and may choose not to devote substantial resources to it.

Imagine what kind of legal power - and money - a company like Sony could bring to bear in a copyright violation. Imagine the legal delays that can be done. Can you afford the plane fare to fly you and your attorney to wherever the lawsuit will be contested?

Please see the FSF's page on Violations of the GPL, LGPL and GFDL, in which it emphasizes that the only person who can act on a violation is the copyright holder.

Note that it may be required, and definitely is advisable, that you assign copyright to the FSF in writing - not in comments in the source code, but as a legal document written on paper with a "wet signature". I believe the FSF has a page about that somewhere but I couldn't find it. I do know that copyrights can only be assigned to someone else in writing.

There is one good reason to reserve copyright to yourself, and it is a considerable reason. You may wish to be able to provide the software under a proprietary license to someone else who wouldn't be required to redistribute source, perhaps for a fee. That is done, for example, for the proprietary license to CygWin, which allows proprietary Unix applications to be readily ported to windows without requiring source code disclosure, in the case that the proprietary license to CygWin is purchased.

If relicensing your source is not in your plans, assign the copyright to the FSF.

Mike

You should be ashamed of yourself, posting such accusations anonymously. If you fired me within a week then you'll be happy to post using your real name. Why don't you post a response with your name? I stand by the quality of my work. My work is of exceedingly high quality and my clients know it.

And no, I'm no karma whore, and I'm not trying to find a consulting gig by posting articles such as this.

Instead, I'm trying to help out the kernel developers who need people testing their software, and to improve the overall quality of Free Software.

The kernel I feel is of very high quality, but it is challenging for the kernel developers to keep it so, because it can be configured with so many variations and has to satisfy so many needs. That's why it needs an especially large amount of testing.

There are many free and open source software products that I feel do not live up to even modest quality standards, and I'm trying to improve the situation by posting educational articles such as the ones on the Linux Quality Database.

Have a look at freshmeat and see how many programs have yet to reach 1.0 status - many of those programs are in every day use on Linux distributions and in fact cause me and other users significant problems. I'm trying to help remedy the situation by providing advice, guidance and pointers to good tools.

Mike

If you'd like to do some manner of methodical testing of the kernel before putting it into production use, or if you'd like to contribute to quality assurance of Linux kernels, please read:

• Using Test Suites to Validate the Linux Kernel

There are also some tests such as Memtest86 that you can use for checking the basic health of your hardware (some of which, like VA Linux' Cerberus, can be quite extreme). Memtest86 found a bad PC133 memory module in one of my machines that consistently passes the BIOS startup memory test.

Suggestions for other test suites, strategies for testing the Linux kernel or other things I can write about on the general topic of quality assurance and better software practice are appreciated.

One test that's not mentioned in the article yet is memtester. It is a user-space program that will run on Linux or other Unix-like OS and will test as much of the memory as it can mlock; I found that I could test about half of my installed memory on a debian PPC macintosh.

Memtest86 is advantageous in that it will test all of the installed memory, but memtester will run on non-PC clone and non-x86 hardware.

Mike

I wrote the Mac version of Sensiva as one of the projects for my consulting business, under contract to Sensiva Inc.. I encourage you to try it out. It's especially useful if you have a graphics tablet but it will work well with a mouse too.

It was quite an interesting program to write.

Mike

I wrote the Mac version of Sensiva as one of the projects for my consulting business, under contract to Sensiva Inc.. I encourage you to try it out. It's especially useful if you have a graphics tablet but it will work well with a mouse too.

It was quite an interesting program to write.

Mike

Yes, I'm "that spellchecker guy", and it's not shareware, it is a commercial product which is still supported.

Spellswell was one of the first commercial products for the BeOS and remains a supported commercial product. I won an honorable mention in the Be Master's Awards for bringing Spellswell to the BeOS.

It wasn't the "path to riches" I was seeking - I knew that a new operating system would need a spellchecker as a standard system service just as you, and I felt that the right thing to do in combating the Microsoft Monopoly was to bring this standard system service to the BeOS because I had access to its source code and Working Software's consent to use it for this purpose.

In choosing to develop on a particular platform you are voting with your brain and the fingers you type with; I was voting for the BeOS with my efforts. I was not trying to get rich; what I didn't expect was to get lied to.

Both Spellswell for the BeOS, and Spellswell for the Mac OS, which was bundled with Eudora, use the Word Services Suite, which allows each of them to communicate with a number of word processors and email clients as if Spellswell were a built-in menu item.

Several other products support Word Services, and I have proposed to bring it to XWindows as well.

The Spellswell bundled with Eudora was not shareware either and Working Software was paid a license fee for each copy of Eudora it was bundled with. I trust you honored your license agreement and did not make unauthorized copies of this supported commercial product.

The difficulties that Working Software has had are typical of the troubles that every developer of commercial software has had as a result of their decision to support the BeOS. But these difficulties stem not from poor quality products, but from trusting folks like Jean-Louis Gassee to live up to their word.

Mike

Yes, I'm "that spellchecker guy", and it's not shareware, it is a commercial product which is still supported.

Spellswell was one of the first commercial products for the BeOS and remains a supported commercial product. I won an honorable mention in the Be Master's Awards for bringing Spellswell to the BeOS.

It wasn't the "path to riches" I was seeking - I knew that a new operating system would need a spellchecker as a standard system service just as you, and I felt that the right thing to do in combating the Microsoft Monopoly was to bring this standard system service to the BeOS because I had access to its source code and Working Software's consent to use it for this purpose.

In choosing to develop on a particular platform you are voting with your brain and the fingers you type with; I was voting for the BeOS with my efforts. I was not trying to get rich; what I didn't expect was to get lied to.

Both Spellswell for the BeOS, and Spellswell for the Mac OS, which was bundled with Eudora, use the Word Services Suite, which allows each of them to communicate with a number of word processors and email clients as if Spellswell were a built-in menu item.

Several other products support Word Services, and I have proposed to bring it to XWindows as well.

The Spellswell bundled with Eudora was not shareware either and Working Software was paid a license fee for each copy of Eudora it was bundled with. I trust you honored your license agreement and did not make unauthorized copies of this supported commercial product.

The difficulties that Working Software has had are typical of the troubles that every developer of commercial software has had as a result of their decision to support the BeOS. But these difficulties stem not from poor quality products, but from trusting folks like Jean-Louis Gassee to live up to their word.

Mike

I have no doubt about the engineering quality on the BeOS, it is one of the finest examples of software engineering I have come across in my career - that's the reason why I was so excited to work with it, and kept with it for so long and still support my application on it (to be updated soon now that my laptop is back from the repair shop).

The work I am doing with the Linux Quality Database is inspired in large part by a desire to bring this kind of quality to Open Source and Free Software systems. I use Linux on a daily basis in my own work now but honestly my experience of it wears me down and I fell so... refreshed whenever I restart into the BeOS.

But it is clear to me that Be, Inc.'s troubles are due to lack of wisdom and commitment among its management despite the best efforts of its engineers. From the start, Be made very little effort to market to the public, even though anyone who ever tried the BeOS immediately liked it and usually wanted to run it on their machine.

After a while it became clear that Be had problems keeping its commitments to its developers. I lost out on a lot of evenings and weekends spent developing a product that didn't sell well when I could have developed another product for the Mac OS - I was well into one for the Mac but gave it up for the BeOS. The investors who funded BeatWare and Adamation lost millions of dollars on their BeOS development efforts; the companies were only saved when they ported their products to the Mac OS and Windows.

Read about my observations about the difficulties that a number of companies have had surviving in the turbulent world of High-Tech, including my advice as to why business partnerships with Be are best avoided:

• The Valley is a Harsh Mistress

Learn how I am working with others to take the power from the hands of the OS vendors and put it back into the hands of the third-party developers and the public:

• Freeing the Developer from OS Vendor Shackles

Read a rather blunt summary posted to Be's developer mailing list about how disappointed I was that the company had failed to live up to the promises it had made to its own developers who had labored hard, and usually with little or no compensation, to bring great applications to the BeOS:

• Some of Us Work for a Living

That was the last message I posted to bedevtalk; the moderator unsubscribed me because he felt it was inappropriate to discuss business matters relevant to BeOS third-party developers on Be's third-party developer mailing list.

Be dug their own grave. If someone comes to their rescue with new funding, I'd like to suggest that the package include a new top-level management team with a mandate to shake things up. Firing the sales-prevention team would help.

I'll be sad if the BeOS dies and I hope they do open source it. It is likely that it uses licensed technology so that they could not legally open source the whole thing; let us hope they take the route netscape did and remove the proprietary parts and allow the open sourcing of the rest, rather than allowing it to die.

Mike

I have no doubt about the engineering quality on the BeOS, it is one of the finest examples of software engineering I have come across in my career - that's the reason why I was so excited to work with it, and kept with it for so long and still support my application on it (to be updated soon now that my laptop is back from the repair shop).

The work I am doing with the Linux Quality Database is inspired in large part by a desire to bring this kind of quality to Open Source and Free Software systems. I use Linux on a daily basis in my own work now but honestly my experience of it wears me down and I fell so... refreshed whenever I restart into the BeOS.

But it is clear to me that Be, Inc.'s troubles are due to lack of wisdom and commitment among its management despite the best efforts of its engineers. From the start, Be made very little effort to market to the public, even though anyone who ever tried the BeOS immediately liked it and usually wanted to run it on their machine.

After a while it became clear that Be had problems keeping its commitments to its developers. I lost out on a lot of evenings and weekends spent developing a product that didn't sell well when I could have developed another product for the Mac OS - I was well into one for the Mac but gave it up for the BeOS. The investors who funded BeatWare and Adamation lost millions of dollars on their BeOS development efforts; the companies were only saved when they ported their products to the Mac OS and Windows.

Read about my observations about the difficulties that a number of companies have had surviving in the turbulent world of High-Tech, including my advice as to why business partnerships with Be are best avoided:

• The Valley is a Harsh Mistress

Learn how I am working with others to take the power from the hands of the OS vendors and put it back into the hands of the third-party developers and the public:

• Freeing the Developer from OS Vendor Shackles

Read a rather blunt summary posted to Be's developer mailing list about how disappointed I was that the company had failed to live up to the promises it had made to its own developers who had labored hard, and usually with little or no compensation, to bring great applications to the BeOS:

• Some of Us Work for a Living

That was the last message I posted to bedevtalk; the moderator unsubscribed me because he felt it was inappropriate to discuss business matters relevant to BeOS third-party developers on Be's third-party developer mailing list.

Be dug their own grave. If someone comes to their rescue with new funding, I'd like to suggest that the package include a new top-level management team with a mandate to shake things up. Firing the sales-prevention team would help.

I'll be sad if the BeOS dies and I hope they do open source it. It is likely that it uses licensed technology so that they could not legally open source the whole thing; let us hope they take the route netscape did and remove the proprietary parts and allow the open sourcing of the rest, rather than allowing it to die.

Mike

I have no doubt about the engineering quality on the BeOS, it is one of the finest examples of software engineering I have come across in my career - that's the reason why I was so excited to work with it, and kept with it for so long and still support my application on it (to be updated soon now that my laptop is back from the repair shop).

The work I am doing with the Linux Quality Database is inspired in large part by a desire to bring this kind of quality to Open Source and Free Software systems. I use Linux on a daily basis in my own work now but honestly my experience of it wears me down and I fell so... refreshed whenever I restart into the BeOS.

But it is clear to me that Be, Inc.'s troubles are due to lack of wisdom and commitment among its management despite the best efforts of its engineers. From the start, Be made very little effort to market to the public, even though anyone who ever tried the BeOS immediately liked it and usually wanted to run it on their machine.

After a while it became clear that Be had problems keeping its commitments to its developers. I lost out on a lot of evenings and weekends spent developing a product that didn't sell well when I could have developed another product for the Mac OS - I was well into one for the Mac but gave it up for the BeOS. The investors who funded BeatWare and Adamation lost millions of dollars on their BeOS development efforts; the companies were only saved when they ported their products to the Mac OS and Windows.

Read about my observations about the difficulties that a number of companies have had surviving in the turbulent world of High-Tech, including my advice as to why business partnerships with Be are best avoided:

• The Valley is a Harsh Mistress

Learn how I am working with others to take the power from the hands of the OS vendors and put it back into the hands of the third-party developers and the public:

• Freeing the Developer from OS Vendor Shackles

Read a rather blunt summary posted to Be's developer mailing list about how disappointed I was that the company had failed to live up to the promises it had made to its own developers who had labored hard, and usually with little or no compensation, to bring great applications to the BeOS:

• Some of Us Work for a Living

That was the last message I posted to bedevtalk; the moderator unsubscribed me because he felt it was inappropriate to discuss business matters relevant to BeOS third-party developers on Be's third-party developer mailing list.

Be dug their own grave. If someone comes to their rescue with new funding, I'd like to suggest that the package include a new top-level management team with a mandate to shake things up. Firing the sales-prevention team would help.

I'll be sad if the BeOS dies and I hope they do open source it. It is likely that it uses licensed technology so that they could not legally open source the whole thing; let us hope they take the route netscape did and remove the proprietary parts and allow the open sourcing of the rest, rather than allowing it to die.

Mike

Are you tired of all those annoying paid search engine placement services? Ever tried using the free ones, only to be annoyed with tons of ads and to find your URL submissions blocked by the robosubmission filters on the search sites?

Well, I'm tired of them too, and I write pages that I submit to search engines from time to time, and I've come up with what I feel is the best way to submit links to a bunch of sites:

Direct links into the pages that have the URL submission forms on a bunch of search engines.

• Search Engine Submission Form Index

Keep a text window open with your URL, title, description, for-public-consumption email address and the like, and use "Open Page in New Window" on all these links to manually copy and paste your information into a bunch of search engine submission forms.

That's it!

I got all these search engines off the Search Engines Category at the Open Directory Project. If you know of any pages that list a bunch of other search engines (there are many smaller ones, and a lot of special purpose ones) then drop me a line at crawford@goingware.com.

In my index I provide brief notes about some of the engines, including mentioning whether they refuse to accept submissions without payment. I don't provide links to submission forms for the engines that won't list a site for free, and I'd like to ask you not to support the trend towards paid index and spider placement.

You should understand that the vast majority of visitors to your sites don't get there through search engines, they get there because other people like your page and give you a link. The main value of search engines is to "prime the pump" so a few people start finding your site and then know to create a link for it.

Create successful web sites by writing good web sites - see Some Web Application Design Basics for links to a few good pages written by experts that will start you well on the road to an appealing, successful website.

Thank you for your attention.

Mike

Are you tired of all those annoying paid search engine placement services? Ever tried using the free ones, only to be annoyed with tons of ads and to find your URL submissions blocked by the robosubmission filters on the search sites?

Well, I'm tired of them too, and I write pages that I submit to search engines from time to time, and I've come up with what I feel is the best way to submit links to a bunch of sites:

Direct links into the pages that have the URL submission forms on a bunch of search engines.

• Search Engine Submission Form Index

Keep a text window open with your URL, title, description, for-public-consumption email address and the like, and use "Open Page in New Window" on all these links to manually copy and paste your information into a bunch of search engine submission forms.

That's it!

I got all these search engines off the Search Engines Category at the Open Directory Project. If you know of any pages that list a bunch of other search engines (there are many smaller ones, and a lot of special purpose ones) then drop me a line at crawford@goingware.com.

In my index I provide brief notes about some of the engines, including mentioning whether they refuse to accept submissions without payment. I don't provide links to submission forms for the engines that won't list a site for free, and I'd like to ask you not to support the trend towards paid index and spider placement.

You should understand that the vast majority of visitors to your sites don't get there through search engines, they get there because other people like your page and give you a link. The main value of search engines is to "prime the pump" so a few people start finding your site and then know to create a link for it.

Create successful web sites by writing good web sites - see Some Web Application Design Basics for links to a few good pages written by experts that will start you well on the road to an appealing, successful website.

Thank you for your attention.

Mike

Are you tired of all those annoying paid search engine placement services? Ever tried using the free ones, only to be annoyed with tons of ads and to find your URL submissions blocked by the robosubmission filters on the search sites?

Well, I'm tired of them too, and I write pages that I submit to search engines from time to time, and I've come up with what I feel is the best way to submit links to a bunch of sites:

Direct links into the pages that have the URL submission forms on a bunch of search engines.

• Search Engine Submission Form Index

Keep a text window open with your URL, title, description, for-public-consumption email address and the like, and use "Open Page in New Window" on all these links to manually copy and paste your information into a bunch of search engine submission forms.

That's it!

I got all these search engines off the Search Engines Category at the Open Directory Project. If you know of any pages that list a bunch of other search engines (there are many smaller ones, and a lot of special purpose ones) then drop me a line at crawford@goingware.com.

In my index I provide brief notes about some of the engines, including mentioning whether they refuse to accept submissions without payment. I don't provide links to submission forms for the engines that won't list a site for free, and I'd like to ask you not to support the trend towards paid index and spider placement.

You should understand that the vast majority of visitors to your sites don't get there through search engines, they get there because other people like your page and give you a link. The main value of search engines is to "prime the pump" so a few people start finding your site and then know to create a link for it.

Create successful web sites by writing good web sites - see Some Web Application Design Basics for links to a few good pages written by experts that will start you well on the road to an appealing, successful website.

Thank you for your attention.

Mike

I should say right up front that although I subscribe to the AbiWord list, the following is definitely my own personal opinion and I have no idea how it might correspond to the opinion of any of the AbiWord developers.

I think doing cross-platform development is of critical importance both to the software developer and the public. Find out why at:

• Freeing the Developer from OS Vendor Shackles

There are a number of cross-platform application frameworks, one of which is the framework AbiWord is built on. Others you may be familiar with are the Mozilla framework and GTK+. The above essay is on the website for the ZooLib cross-platform application framework.

You can find a list of many application frameworks in several languages, many of which are cross-platform, and many of which are free or open source, at the GUI Toolkit, Framework page.

Mike

Ever tried to use one of those godforsaken URL submission applications, the ones that take your page and spams a bunch of spiders with it?

Or heaven forbid, have you paid for their paid search engine submission packages?

Wouldn't it be nice if somebody just collected links to all the "Add URL" pages for the engines in one place so you could quickly do manual submissions? If you just had links to click on, and you had a text window open that you could copy and paste your info from (URL, description, email), then you could do manual submission much quicker than using one of these applications.

And you won't run into the barriers the search engines put up against URL submission bots.

Here you go:

• Search Engine Submission Form Index

Leave your tip in the jar.

Mike

Ever tried to use one of those godforsaken URL submission applications, the ones that take your page and spams a bunch of spiders with it?

Or heaven forbid, have you paid for their paid search engine submission packages?

Wouldn't it be nice if somebody just collected links to all the "Add URL" pages for the engines in one place so you could quickly do manual submissions? If you just had links to click on, and you had a text window open that you could copy and paste your info from (URL, description, email), then you could do manual submission much quicker than using one of these applications.

And you won't run into the barriers the search engines put up against URL submission bots.

Here you go:

• Search Engine Submission Form Index

Leave your tip in the jar.

Mike

Hot off the presses tonight:

• Use Validators and Load Generators to Test Your Web Applications

Maybe the folks who write the Slashcode would find it helpful.

I've posted this here before, but don't want the IBM folks who might be reading to miss it:

• Using Test Suites to Validate the Linux Kernel

Comments, criticism, additional links and resources to add, suggestions for future articles to write and of course articles you would like to write are appreciated.

I could also use some help from someone with expertise in designing database schemas.

Thank you for your attention.

Mike

I spent a summer working at CERN doing my senior thesis for a B.A. degree in Physics at UC Santa Cruz (I was working with the Spin Muon Collaboration, mostly working on data analysis software - I made the mistake of walking down the hall at the physics department asking each professor I met if they could use an experienced software engineer who needed a thesis topic! Mmmm... FORTRAN.).

While I was there I noticed that the CERN neutrino beam went right down the main street of the nearby town of St. Genis in France and on into the Jura Mountains. I wonder if the townspeople in St. Genis would feel comfortable knowing they were being irradiated, even if they understood the particles wouldn't interact.

You see, while the detector that's the subject of this story detects neutrinos of cosmic origin, you can also make them artificially, and with controlled energies and other desirable characteristics, by shooting a high energy particle beam into one end of a long pile of dirt.

The particles shower but are then absorbed by the dirt - except for the neutrinos produced by the showers. Enough dirt, and whatever comes out the other end is pretty much pure neutrino beam.

If you put in an intermediate amount of shielding, you get a mix of muons and neutrinos.

The way you detect these artificial particle beams is typically with packs filled with photographic film sealed in a dark chamber. Just beam it for a while and every zillionth particle will leave a little speck on some of the film.

Ever heard of neutrino oscillations? They proposed the theory to explain the lack of expected neutrino flux in one of the earlier underground neutrino detectors. It takes 10,000 years for heat from the center of the Sun to convect to the surface before it can shine directly on the earth, but neutrinos radiate from the core to the earth in 8 minutes because they don't interact.

Only problem is, we weren't getting many neutrinos. The first suspicion was that the Sun had begun to die but the cooling part of the interior hadn't reach the surface yet - that is, we hadn't visibly received the bad news but had found out ahead of time with the neutrino detector.

If neutrinos change identities into types that a given sensor is not sensitive to, though, it would explain this. But for this to be the case, the neutrino would have to have a very small, but non-zero mass. It's been the work of decades to try to measure this mass.

In the particle beam at CERN they would measure the neutrino flux at different points along the beam to see if they got more and less intense as they oscillated between electron, muon and tau neutrinos.

Enjoy!

Mike

That doesn't count (although it does present an argument for why it's better to set up Mom with an OS that supports remote administration). Let's pretend you're calling in from a payphone from a noisy nightclub in Berlin because you got voicemail from Mom that her Linux box won't work anymore, or some new software which she did correctly install won't work until a patch is applied to some OS component.

Now talk her through it over the phone with no feeback from her machine but what Mom reads you over the phone - I used to help people configure and link Microport System V/AT kernels over the phone all the time when I worked in their tech support department.

The command line does have its advantages over GUI in purely voice tech support but it's no substitute for just having nothing ever go wrong at all.

While difficult to achieve, complete reliability and ease of use are technical possibilities.

Mike

That doesn't count (although it does present an argument for why it's better to set up Mom with an OS that supports remote administration). Let's pretend you're calling in from a payphone from a noisy nightclub in Berlin because you got voicemail from Mom that her Linux box won't work anymore, or some new software which she did correctly install won't work until a patch is applied to some OS component.

Now talk her through it over the phone with no feeback from her machine but what Mom reads you over the phone - I used to help people configure and link Microport System V/AT kernels over the phone all the time when I worked in their tech support department.

The command line does have its advantages over GUI in purely voice tech support but it's no substitute for just having nothing ever go wrong at all.

While difficult to achieve, complete reliability and ease of use are technical possibilities.

Mike

Ever helped your mom fix her PPP connection on Windows or Mac OS? Did you do it in person, or over the phone?

Now help your mom download, apply, configure, compile and install a source code patch!

I know people who are blessed with parents who know how to program, but my mom wanted to go back to CompuServe after Dad fixed her up with Earthlink because earthlink gave them netscape on the installation CD to read her email with.

She didn't know, until I came to visit, downloaded and tested every email client available for the Mac OS, and set her up with Sono Software's Musashi that everyone's email wasn't hardwired to their ISP. I tried to explain to her before I visited that she could get a different mail program, but didn't understand - she didn't know she was running Netscape to read her email, she thought she was reading her email in Earthlink.

My dad knows a little programming - FORTRAN IV on an IBM 360 via punchcards.

If something breaks on their little Mac (and they manage to do it somehow), it stays broken until their son the software consultant flies home to visit.

Please contribute articles to the Linux Quality Database on how to write quality software, test it, or send in links to quality information and resources that I can link to from the page or write articles about.

Especially important are articles that are accessible to ordinary users or inexperienced programers on how they can contribute in a meaningful way to QA testing of the Free Software product of their choice. An example is my article (still in progress) Using Test Suites to Validate the Linux Kernel.

Planning for the kernel testing database that I have in mind is still in a vaporware phase. If you know about designing database schemas I could really use your help in architecting this thing. I could write the database access code myself but I don't want to plan the schema because I'm sure I'd make a mess of it. I've never designed a real schema before.

I'd like to use Enhydra as the application server - it's pretty good and open source. You can integrate it with Apache or the Netscape server, or it comes with its own Java HTTP server built-in that you can use either for testing or production.

When you use the Enhydra multiserver in development the whole enhydra system can run in one Java VM process. I echoed out the java command line from the startup script to find out what all the parameters were then put the command line into NetComputing's AnyJ debugger (free as in beer for Linux) to run a whole web application in a source debugger. Much nicer and more effective than printing "here I am" to the console.

I believe they have Enhydra running under Kaffe now. It didn't when I tried it but the startup scripts now support it as an option. I reported the problem to the Kaffe folks quite a while ago and I guess they fixed it.

Mike

Ever helped your mom fix her PPP connection on Windows or Mac OS? Did you do it in person, or over the phone?

Now help your mom download, apply, configure, compile and install a source code patch!

I know people who are blessed with parents who know how to program, but my mom wanted to go back to CompuServe after Dad fixed her up with Earthlink because earthlink gave them netscape on the installation CD to read her email with.

She didn't know, until I came to visit, downloaded and tested every email client available for the Mac OS, and set her up with Sono Software's Musashi that everyone's email wasn't hardwired to their ISP. I tried to explain to her before I visited that she could get a different mail program, but didn't understand - she didn't know she was running Netscape to read her email, she thought she was reading her email in Earthlink.

My dad knows a little programming - FORTRAN IV on an IBM 360 via punchcards.

If something breaks on their little Mac (and they manage to do it somehow), it stays broken until their son the software consultant flies home to visit.

Please contribute articles to the Linux Quality Database on how to write quality software, test it, or send in links to quality information and resources that I can link to from the page or write articles about.

Especially important are articles that are accessible to ordinary users or inexperienced programers on how they can contribute in a meaningful way to QA testing of the Free Software product of their choice. An example is my article (still in progress) Using Test Suites to Validate the Linux Kernel.

Planning for the kernel testing database that I have in mind is still in a vaporware phase. If you know about designing database schemas I could really use your help in architecting this thing. I could write the database access code myself but I don't want to plan the schema because I'm sure I'd make a mess of it. I've never designed a real schema before.

I'd like to use Enhydra as the application server - it's pretty good and open source. You can integrate it with Apache or the Netscape server, or it comes with its own Java HTTP server built-in that you can use either for testing or production.

When you use the Enhydra multiserver in development the whole enhydra system can run in one Java VM process. I echoed out the java command line from the startup script to find out what all the parameters were then put the command line into NetComputing's AnyJ debugger (free as in beer for Linux) to run a whole web application in a source debugger. Much nicer and more effective than printing "here I am" to the console.

I believe they have Enhydra running under Kaffe now. It didn't when I tried it but the startup scripts now support it as an option. I reported the problem to the Kaffe folks quite a while ago and I guess they fixed it.

Mike

Ever helped your mom fix her PPP connection on Windows or Mac OS? Did you do it in person, or over the phone?

Now help your mom download, apply, configure, compile and install a source code patch!

I know people who are blessed with parents who know how to program, but my mom wanted to go back to CompuServe after Dad fixed her up with Earthlink because earthlink gave them netscape on the installation CD to read her email with.

She didn't know, until I came to visit, downloaded and tested every email client available for the Mac OS, and set her up with Sono Software's Musashi that everyone's email wasn't hardwired to their ISP. I tried to explain to her before I visited that she could get a different mail program, but didn't understand - she didn't know she was running Netscape to read her email, she thought she was reading her email in Earthlink.

My dad knows a little programming - FORTRAN IV on an IBM 360 via punchcards.

If something breaks on their little Mac (and they manage to do it somehow), it stays broken until their son the software consultant flies home to visit.

Please contribute articles to the Linux Quality Database on how to write quality software, test it, or send in links to quality information and resources that I can link to from the page or write articles about.

Especially important are articles that are accessible to ordinary users or inexperienced programers on how they can contribute in a meaningful way to QA testing of the Free Software product of their choice. An example is my article (still in progress) Using Test Suites to Validate the Linux Kernel.

Planning for the kernel testing database that I have in mind is still in a vaporware phase. If you know about designing database schemas I could really use your help in architecting this thing. I could write the database access code myself but I don't want to plan the schema because I'm sure I'd make a mess of it. I've never designed a real schema before.

I'd like to use Enhydra as the application server - it's pretty good and open source. You can integrate it with Apache or the Netscape server, or it comes with its own Java HTTP server built-in that you can use either for testing or production.

When you use the Enhydra multiserver in development the whole enhydra system can run in one Java VM process. I echoed out the java command line from the startup script to find out what all the parameters were then put the command line into NetComputing's AnyJ debugger (free as in beer for Linux) to run a whole web application in a source debugger. Much nicer and more effective than printing "here I am" to the console.

I believe they have Enhydra running under Kaffe now. It didn't when I tried it but the startup scripts now support it as an option. I reported the problem to the Kaffe folks quite a while ago and I guess they fixed it.

Mike