6 Mar
2007
6 Mar
'07
10:12 a.m.
I suppose (or is it like that already?) bitsavers
needs some form of database
This is difficult to do the way that it is hosted and mirrored currently. Right
now, it is just a directory heirarchy with a couple of files I have to manually
edit and upload every time I add something (IndexByDate and Whatsnew in the pdf
directory) There are no equivalent files in the 'bits' directory.
I know Jay would like it to distribute the transfer load across the mirrors, too.
6 Mar
6 Mar
10:24 a.m.
Al wrote...
I know Jay would like it to distribute the transfer
load across the
mirrors, too.
I've been working on that. The code I was using seems to have some issues so
I started looking for a different way.
I've also been playing with RSS and will have something along those lines at
some point.
Jay
11:34 a.m.
Jay West wrote:
Al wrote...
Do we have any transfer statistics for the mirrors at the moment?
bitsavers.vt100.net currently serves 300 GiB per month on average.
I know Jay would like it to distribute the
transfer load across the
mirrors, too.
I've been working on that. The code I was using seems to have some
issues so I started looking for a different way. 
10:54 a.m.
New subject: Data I/O IX
Hi
There is a Data I/O IX on ebay. It doesn't look like it would take
a Uni-pack. Does anyone know what PROM/EPROM types this machine will
program? Is it able to do 1702A's ?
Dwight
_________________________________________________________________
Mortgage rates as low as 4.625% - Refinance $150,000 loan for $579 a month.
Intro*Terms
https://www2.nextag.com/goto.jsp?product=100000035&url=%2fst.jsp&tm…
8 Mar
8 Mar
3:46 p.m.
New subject: Data I/O IX
dwight elvey wrote:
Hi
There is a Data I/O IX on ebay. It doesn't look like it would take
a Uni-pack. Does anyone know what PROM/EPROM types this machine will
program? Is it able to do 1702A's ?
Dwight
Or build your own ... I am reading the book
"how to build your own working microcomputer" and they use
a 8080, and 1708's and 2102's for memory.
This is build it from scratch book.
Ben alias woodelf
PS has anybody ever built the computer in the book?
3:59 p.m.
New subject: Data I/O IX
On Mar 8, 2007, at 3:46 PM, woodelf wrote:
I've not built the computer, but I am considering it. I obtained
a copy of the book for a few dollars on eBay last summer.
-Dave
--
Dave McGuire
Port Charlotte, FL
program? Is it
able to do 1702A's ?
Dwight
Or build your own ... I am reading the book
"how to build your own working microcomputer" and they use
a 8080, and 1708's and 2102's for memory.
This is build it from scratch book.
Ben alias woodelf
PS has anybody ever built the computer in the book? 
6 Mar
6 Mar
2:18 p.m.
Al Kossow wrote:
Yes, I was originally thinking that in the current situation some PHP code
could do the work of the webserver's directory listing code and incorporate
the description field from Whatsnew.txt when people view the site via HTTP.
However, looking at the file it's 278KB in size and over 4000 lines - I
suspect the burden of searching that file to match file paths to descriptions
might be a little too great.
Some sort of background process to run a couple of times a day and update
per-directory equivalents of Whatsnew.txt could work I suppose; any "dir
listing" code will then only need to operate on a small file. Such an approach
"feels" messy though, but might be the only option without a full db.
Either that or status quo is maintained and users have to explicitly look at
Whatsnew.txt. to find the descriptions for the directory listings that they're
looking at.
cheers
Jules
I suppose (or
is it like that already?) bitsavers needs some form of
database
This is difficult to do the way that it is hosted and mirrored
currently. Right
now, it is just a directory heirarchy with a couple of files I have to
manually
edit and upload every time I add something (IndexByDate and Whatsnew in
the pdf
directory) There are no equivalent files in the 'bits' directory. 
2:35 p.m.
Rumor has it that Jules Richardson may have mentioned these words:
Yes, I was originally thinking that in the current
situation some PHP code
could do the work of the webserver's directory listing code and
incorporate the description field from Whatsnew.txt when people view the
site via HTTP. However, looking at the file it's 278KB in size and over
4000 lines - I suspect the burden of searching that file to match file
paths to descriptions might be a little too great.
If it's a file that's searched often, but updated rarely (say, once/dayish)
a CDB database might be just the ticket... it's a single-hit flat-file
lookup system. I *think* there's PHP modules for it; I know there's Perl &
Python modules for it. So for something that's searched often, but updated
kinda rarely, they don't add a lot of load to the system and might be "just
the ticket."
HTH,
Roger "Merch" Merchberger
--
Roger "Merch" Merchberger -- SysAdmin, Iceberg Computers
zmerch at 30below.com
What do you do when Life gives you lemons,
and you don't *like* lemonade?????????????
3:05 p.m.
It was thus said that the Great Jules Richardson once stated:
Al Kossow wrote:
Yes, I was originally thinking that in the current situation some PHP code
could do the work of the webserver's directory listing code and incorporate
the description field from Whatsnew.txt when people view the site via HTTP.
However, looking at the file it's 278KB in size and over 4000 lines - I
suspect the burden of searching that file to match file paths to
descriptions might be a little too great.
Some sort of background process to run a couple of times a day and update
per-directory equivalents of Whatsnew.txt could work I suppose; any "dir
listing" code will then only need to operate on a small file. Such an
approach "feels" messy though, but might be the only option without a full
db.
Apache has the ability to associate descriptions to filenames for the
directory indexes it can produce (if you don't have an index.html file, and
you allow directory listings to be produced). Look up the "AddDescription"
directive, and such directives can be added to a .htaccess file in each
directory.
-spc (And why does it feel messy?)
I suppose
(or is it like that already?) bitsavers needs some form of database
This is difficult to do the way that it is hosted and mirrored currently. Right
now, it is just a directory heirarchy with a couple of files I have to manually
edit and upload every time I add something (IndexByDate and Whatsnew in the pdf
directory) There are no equivalent files in the 'bits' directory.
4:11 p.m.
Sean Conner wrote:
Is it customisable as to where and how it pulls the description information of
of a file for inclusion, though?
Some sort of
background process to run a couple of times a day and update
per-directory equivalents of Whatsnew.txt could work I suppose; any "dir
listing" code will then only need to operate on a small file. Such an
approach "feels" messy though, but might be the only option without a full
db.
Apache has the ability to associate descriptions to filenames for the
directory indexes it can produce (if you don't have an index.html file, and
you allow directory listings to be produced). Look up the "AddDescription"
directive, and such directives can be added to a .htaccess file in each
directory. -spc (And why does it feel messy?)
Only because the same information's essentially being maintained in two places...
4:27 p.m.
I note small problem in the bitsavers archive that may deserve some
thought. Documentation for the IBM Selectric I/O writer is located
in the CDC directory. Why? Because the documentation originated at
CDI, not IBM, even if it describes an IBM product.
Is there a way to symbolically link the CDI documents into the IBM
directory? Are there any other instances of this sort of thing?
Not particularly germane to the RSS feed thing, but something I've
wondered about.
Cheers,
Chuck
4:48 p.m.
It was thus said that the Great Jules Richardson once stated:
Sean Conner wrote:
Is it customisable as to where and how it pulls the description information
of of a file for inclusion, though?
Not really. It looks like:
AddDescription "A picture of a PDP-11" pdp11.jpg
AddDescription "Theory of Operations Guide for a PDP-11" php11.pdf
But it saves having to flip back and forth between the listing file and
the file listing ... um ... if that makes any sense.
-spc (Used this for one if his sites ... )
Some sort
of background process to run a couple of times a day and update
per-directory equivalents of Whatsnew.txt could work I suppose; any "dir
listing" code will then only need to operate on a small file. Such an
approach "feels" messy though, but might be the only option without a
full db.
Apache has the ability to associate descriptions to filenames for the
directory indexes it can produce (if you don't have an index.html file, and
you allow directory listings to be produced). Look up the "AddDescription"
directive, and such directives can be added to a .htaccess file in each
directory. 
5:30 p.m.
On 06/03/07, Sean Conner <spc at conman.org> wrote:
It strikes me that this is fairly straightforward to create however; a
few lines of php to add it to an .htaccess in each folder when logging
in a new file to that folder. And it should be each to ensure it
gets echoed across the mirrors automatically, if it wouldn't already!
Back on the RSS feed; I've done similar work in the past that reads
directory structures to find new files, to build up a database.
(images, in my case, automatically adding products to an oscommerce
database when it finds a new one.) Again, it should be
straightforward to have a periodic task that scans folders for changes
(just a check on the last-modified time since last scan) grabs the
description from the .htaccess file above, and appends to an RSS file
and even the what's new file if you want.
(I still believe that computers are there to save us work - you should
only have to upload a file, give it a description, and everything else
should propogates to wherever you need it automatically.)
Is it
customisable as to where and how it pulls the description information
of of a file for inclusion, though?
Not really. It looks like:
AddDescription "A picture of a PDP-11" pdp11.jpg
AddDescription "Theory of Operations Guide for a PDP-11" php11.pdf
4:14 p.m.
Jay West wrote:
Jules wrote...
Hmmm... if it's also available at all of the mirror sites then maybe that's an
option.
I guess the trick is to not make any more work for Al, though - he's probably
busy enough as it is :-) In other words, the "publishing" interface would
perhaps have to stay as it is now, but some running process would migrate the
info that he puts into Whatsnew.txt into the db.
an approach "feels" messy though, but
might be the only option without
a full db.
Well, mySQL is already on the box.....
6341
days inactive
6343
days old
14 comments
11 participants
participants (11)
-
aek＠bitsavers.org -
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
dkelvey＠hotmail.com -
julesrichardsonuk＠yahoo.co.uk -
jwest＠classiccmp.org -
mcguire＠neurotica.com -
paul＠frixxon.co.uk -
robert＠irrelevant.com -
spc＠conman.org -
zmerch-cctalk＠30below.com