Accuracy vs. Precision

[ ... ]

Feel free to use it if it works for you. I can hardly claim ownership of the idea. IIFC, I was first exposed to the idea when I was in the 8th grade (the junior high had sort of folded over to the building that the high school was in, while the elementary school had grown a new building.) I do remember the room I was in when I first heard it, which is why I can tell you what grade I was in at the time. :-)

I've seen scientific publications which put the '.' at mid height instead of along the bottom line as we do it -- which made it a little easier to interpret it as different.

:-)

As long as there are enough grouping separators, or a number of digits at the end other than two or three, I can usually handle it without having to annotate the document. After all -- a typical mis-reading suddenly shifts the workpiece to something which would not fit on my machines. :-)

Aha -- knowing what search string to use makes a difference. :-)

Hmm ... so we can blame it on the French? :-)

I'm sorry to hear that.

What is done in scientific circles?

Yes -- I suspect that not many take my viewpoint that the "long scale" is more logical and thus would accept it even though it is not common in my country. :-)

[ ... ]

They have -- differently in different countries, because there are so many different "powers that be". :-)

Thanks, DoN.

Reply to
DoN. Nichols
Loading thread data ...

Yes -- I would be tempted to use it myself -- except that with my continued encouragement to use only plain ASCII not extended ASCII, it would make me a poor example of what I preach. :-)

I consider it to be the right thing to do, to keep the "Subject: " header from continuing to grow as people update it to track subject drift (and maybe people will do so more often?) :-)

Enjoy, DoN.

Reply to
DoN. Nichols

How well does SunOS do UTF-8? These days you can do it with a subscript 0: D?N. or D?N?

Reply to
Steve Ackman

Since the A in ASCII means American, and the Usenet "audience" is wider than that, maybe it's time to upgrade to UTF?

Reply to
Steve Ackman

Actually some of the Usenet uses fonts that are not ASCII. You are simply not members.

From time to time, I get middle east fonts from several cultures to those of Chinese and Japanese. And we also speak Canadian English using general code. But the double byte fonts of some nations make it interesting.

Martin

Mart> >

----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= - Total Privacy via Encryption =---

Reply to
Martin H. Eastburn

Well ... it is Solaris these days. SunOs (4.1.4 was the last) was a BSD based system, and Solaris (starting with Solaris 2.0) was a SysV based system. Solaris 1.* was used as an alternate title for the last few versions of SunOs merged with the X11.

I use iso-8859-15 as my systemwide character encoding, but avoid using anything with the parity bit set (extended ASCII) since it does not come out the same on all systems. FWIW -- both of your examples there came out with a '?' between the 'D' and the 'N', so they must be codes falling in the control codes zone of the lower 7-bit ASCII -- except that the parity bit is set instead of clear. This system (at least with my selected character encodeing) refuses to print any of those 32 characters even when it prints the rest of the extended set nicely.

Maybe if I switched over to Gnome instead of CDE it would show up -- but I still could not trust everyone else to see it clearly.

If I want to make it look as I want, I do it in troff and print to PostScript or PDF where full control of character size is available. :-)

Thanks, DoN.

Reply to
DoN. Nichols

This would still be a problem for anyone using plain terminals for usenet (and yes, I know a few). They don't even have proportional spaced characters available, let alone alternate charactersets.

I even still have a couple of good terminals on hand, though they tend to be used for diagnostics more than anything else these days.

Enjoy, DoN.

Reply to
DoN. Nichols

Some of us remember Trascii or truncated Ascii - 6 bit not 7 or 8 or 16.

Martin

Mart> >> >>

----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= - Total Privacy via Encryption =---

Reply to
Martin H. Eastburn

Yep -- upper-case only, such as was found in the Teletype ASR-33. It resulted in some strange things in the tty driver in early unix boxen. Things like if you logged in in all upper case, it would assume that you were on a an ASR-33 and convert all lower case to upper case, and convert all input to lower case (since the commands were normally in lower case, and it was a case-sensitive system). If you actually *wanted* a sequence of upper-case characters, IIRC, you hit a '^' first, then typed the characters which you wanted to be upper case, and then hit the '^' again. There was another character which I have forgotten which caused the single character following it to be kept as upper case.

I also remember that for a while my ADM-3a terminal was upper-case only -- until I added some RAM chips and an extra character generator ROM. It was also only 12 lines until I added more RAM. :-)

IRRC, it took six RAM chips (an extra bank) to convert it to

24-lines, and an extra RAM chip in each bank to convert it to mixed case.

Enjoy, DoN.

Reply to
DoN. Nichols

I scrapped a bunch of those Lear Siegler ADM-3A terminals for the nice power transformers.

Reply to
Michael A. Terrell

I gave mine to a friend who was using one in a archive and recondition work - Gave him not only the detailed specs but also the Ball manual - it was the tube and high voltage / yoke section - OEM from Ball.

I also upgraded the memory and got the lower case ROM. Mine was a 3A, I worked with a company using 3A's and 3's in a production mode. I used mine to provide a hot spare from time to time. Soroc terminals were popular and then the large expensive and beautiful - Fox and Owl 9x12 units that had Upper and Lower Ascii in two fonts and descending letters (like g that goes below the line. The 70's were interesting

My ASR-33KSR went to a friend who was completing his 'home brew' PDP/LSI-11 that was done on a private chip set and needed a command console.(3 years ago) Mine was still mounted to the shipping pallet and included a box of paper tape and another box of paper.

Martin

Mart> "D>>> Some of us remember Trascii or truncated Ascii - 6 bit not 7 or 8 or 16.

----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= - Total Privacy via Encryption =---

Reply to
Martin H. Eastburn

I have a couple dozen green TTL monitors, Ball, Clinton, Motorola, Zenith. I have mostly 12" , but a couple 9". I don't know if i have any of the old 4" NCR version from their ATMs left, but I had a couple dozen.

I have a bunch (About 18) of National Semiconductor memory cards left. I think they were for the VAX.

Reply to
Michael A. Terrell

Yes -- I had the full manual with schematics, because I got mine as a kit. It also included the schematics and specs for the monitor subassembly from Ball.

The kit brought it into just barely affordable for me as an individual at the time -- and not too much later they were selling assembled for less than that. :-)

Indeed so.

A virgin Teletype. I don't think that the "KSR" belongs in that device name, however. That stands for "Keyboard Send Receive" (no paper tape punch and reader), while the ASR which you had at the start was "Automatic Send Receive", which was implemented by the paper tape punch and reader.

For a while I had a "portable" KSR-33 -- with a companion modem

-- each in moulded carrying cases. I had to do some tricks to make it work as a printer on my Altair 680b computer -- interrupt the receive loop from modem to printer with a reed relay controlled by the RS-232 output from the computer.

Enjoy, DoN.

Reply to
DoN. Nichols

I was just going by the OS your newsreader reports. I'm sure you could change the User-Agent string if you wanted, though there is already a patch out these days that allows "set custom_os" in .slrnrc. The string after slrn-0.9.9 can be whatever you want.

I use iso-8859-1 on the FreeBSD 7.0 box and UTF-8 on the Debian Testing box, which is the "desktop" machine.

On my FreeBSD slrn as well.

The subscript 0 isn't in any of the 8-bit character sets AFAIK. You'd have to go to a UCS to display it, so simply useing Gnome wouldn't do it. See, for instance,

formatting link
You *can* however, see most of the multi-byte characters on gaggle groups.

Reply to
Steve Ackman

O.K. That comes from the data returned by "uname -a", which is:

====================================================================== SunOS Katana 5.10 Generic_120011-14 sun4u sparc SUNW,Sun-Blade-1000 ======================================================================

Where if you take the "SunOs" and the "5.10" it is translated to "Solaris" (SysV based) by those familiar with Suns. If it were "4.??" instead, it would be truly the BSD flavored "SunOs".

Hardly seems worth the trobule. :-)

O.K.

Which sould be sufficient reason to avoid them in postings to usenet.

O.K.

Hmm ... reading that, each character has its own "unique name". I wonder whether that name can be represented in 7-bit ASCII? :-)

No thank you.

As it is, I feel free to let my spam filters use charactersets like "koi8-r" and "Windows-1251" as clues that it is spam, since I have never gotten a *real* e-mail in those (that I can tell, at least). :-)

Enjoy, DoN.

Reply to
DoN. Nichols

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.