FORTH DIMENSIONS Back issues
Michael
Forth Dimensions from 1985 - 1992. All are in good condition. If
anyone is interested, drop me a note and I'll send a full inventory.
I'll accept any reasonable offer because if I don't find a home for
them, I'll probably toss them.
-Mike
John Passaniti
If you don't receive a reasonable offer, I would be happy to save them
from the garbage, scan/OCR them, and place them online for others to
enjoy.
Rod Pemberton
news:6e8c9b5f-0ea6-4d93...@p32g2000vbi.googlegroups.com...
On Dec 31, 12:50 pm, Michael <mfel...@alumni.brown.edu> wrote:
> > Are there any collectors out there? I have about 4 years worth of
> > Forth Dimensions from 1985 - 1992. All are in good condition. If
> > anyone is interested, drop me a note and I'll send a full inventory.
> > I'll accept any reasonable offer because if I don't find a home for
> > them, I'll probably toss them.
>
> from the garbage, scan/OCR them, and place them online for others to
> enjoy.
>
Aren't scanned Forth Dimensions issues already available on forth.org? This
is the link I have.
http://www.forth.org/fd/contents.html
It seems they have from June/July 1978 (Issue 1, vol. 1) to May/August 1999
(Issue 21, vol. 2) as .pdf scanned images (not OCR'd, not text searchable).
Except for Issue #21, forth.org has six volumes per issue (bi-monthly?).
Well, they're there - if or when you're able to connect to forth.org's
intermittent website...
Rod Pemberton
John Passaniti
> It seems they have from June/July 1978 (Issue 1, vol. 1) to May/August 1999
> (Issue 21, vol. 2) as .pdf scanned images (not OCR'd, not text searchable).
> Except for Issue #21, forth.org has six volumes per issue (bi-monthly?).
> Well, they're there - if or when you're able to connect to forth.org's
> intermittent website...
Exactly. I just tried the link you provided, and it wasn't up. And
when it has been up in the past, it was only scanned images so was
only useful if you knew what issue had the content you were looking
for.
But even if it was a duplicated effort and even if forth.org had 100%
uptime, redundancy is a good thing on the web.
Rod Pemberton
> On Jan 1, 4:14 am, "Rod Pemberton" <do_not_h...@nohavenot.cmm> wrote:
> > It seems they have from June/July 1978 (Issue 1, vol. 1) to May/August
1999
> > (Issue 21, vol. 2) as .pdf scanned images (not OCR'd, not text
searchable).
> > Except for Issue #21, forth.org has six volumes per issue (bi-monthly?).
> > Well, they're there - if or when you're able to connect to forth.org's
> > intermittent website...
>
> Exactly. I just tried the link you provided, and it wasn't up.
>
Up now...
> And
> when it has been up in the past, it was only scanned images so was
> only useful if you knew what issue had the content you were looking
> for.
They have a partial index as a .doc somewhere. I didn't see it listed on
their website but found it via Yahoo.
After the first few years, each issue has a master index in volume #6. I
haven't reviewed these.
I've been slowly working through indexing the articles in the main index (so
far, just for personal use...). Unfortunately, the first few years have
*numerous* articles that aren't listed in the main magazine index. It
appears that some of the later ones do also. I haven't begun to index
them...
> But even if it was a duplicated effort and even if forth.org had 100%
> uptime, redundancy is a good thing on the web.
Do they need to be .zip'd (.bzip2 ? etc.) and put up on a file share website
(e.g., RapidShare)? Is this allowed? I haven't checked. I'm not looking
for trouble here. There are 121 files 905MB without compression.
RP
John Passaniti
> > Exactly. I just tried the link you provided, and it wasn't up.
>
> Up now...
It's up. It's down. It would make a great cat toy.
> > But even if it was a duplicated effort and even if forth.org had 100%
> > uptime, redundancy is a good thing on the web.
>
> Do they need to be .zip'd (.bzip2 ? etc.) and put up on a file share website
> (e.g., RapidShare)? Is this allowed? I haven't checked. I'm not looking
> for trouble here. There are 121 files 905MB without compression.
One of the cute things about comp.lang.forth is how many people here
have no idea how inexpensive it is to put content online. Using a
volume-based web hosting service (such as 1&1), you can get 10
gigabytes of disk space for content, a domain name, and even a basic
content management system for $4/month on a shared hosting system. I
spend more than that on lunch. And while such systems have limits,
only the most optimistic would think that the combined weight of the
entire Forth community on such a site would amount to a fraction of
any of those limits.
For those who need more, my current favorite service provider
(Slicehost) offers a virtual server that can be configured with
several flavors of Linux. You get full root access (including to a
console), 10 gigabytes of storage, 100GB of bandwidth for $20/month.
And it's easy to upgrade too, if you need more. I haven't bothered to
look for cheaper, because I like the service, but yeah, you can
probably find cheaper.
121 files and 905MB of content? Trivial, and cheap. Hell, you could
even do it for *free*. Go over to Wordpress.com and set up a blog.
You get 3 gigabytes of disk space, but more importantly, you get very
good system uptime. Structure the content as blog posts, the body
being a full-text index, attachments being the actual content. Come
up with a system of tags to create a taxonomy of the content. Allow
comments for people to make relevant notes. Presto.
winston...@yahoo.com
Something that small you could set up your own ftp server for free on
your own system and use dyndns.org for the dns resolution...
Rod Pemberton
news:74027d4a-ef96-4fbb...@26g2000yqo.googlegroups.com...
Um... Why bother with either suggestion when there is RapidShare?
http://www.rapidshare.com/
Zip it. Upload it. Free storage. Free download. They provide a download
link for it. Send the link to whomever... It's easiest file transfer
option I've seen.
RP
John Passaniti
<winston19842...@yahoo.com> wrote:
> Something that small you could set up your own ftp server for free on
> your own system and use dyndns.org for the dns resolution...
Sure, but that has at least four problems:
1. Most people who would consider such an option have asymmetric
broadband connections. Where I live, standard cable data rates are
10Mbps down and 384kbps up, with real-world speeds below that. So
presuming my data rates are somewhat typical, I could serve roughly
49k bytes per second from my home. So every megabyte transmitted
would take roughly 22 seconds, assuming a single inbound request, not
in competition for any other inbound requests, and with no competition
from outbound requests within my home.
2. The complaint is that forth.org is unreliable. And while I can
say that the computers in my home are reasonably reliable, I can't say
the same for the local broadband service providers. Some of it isn't
their fault (squirrels love chewing on phone and cable lines), but
quite a lot of it is (we just had an outage two days ago that lasted
three hours, the week before, a router went down and wasn't fixed for
two hours). I have no idea how reliable other people's broadband
services are, but that becomes a factor, along with the reliability of
the systems one would set up to grab files from.
3. Most broadband services disallow operating servers. People do it
anyway (me for one), and practically it isn't a problem because most
people do so for personal reasons (in my case, remote access to my
systems when I'm not at home) and take up very little bandwidth.
Operating a server (especially a FTP server) is a great way to get on
the radar screens of system administrators. I wouldn't feel
comfortable defending why I had dozens of connections from all over
slamming a server that I'm not supposed to be operating in the first
place.
4. Nearly every broadband service provider also provides space for
personal web pages. Where I live, this is 2 gigabytes of disk space.
Assuming that is more common than not, why would anyone want to serve
content from home, suffer through a diminished upstream, have to
maintain a reliable server for others, violate the TOS, and be subject
to the slings and arrows of outrageous connectivity... when they could
just slap that content on the personal web space and be done with it?
Operating a FTP server from one's home has some old-school geek-cred
charm. Those who missed out on the first wave of nerd thrills (say,
1995) can recapture that magic time, and gain some useful skills in
the process. The rest of us who live in the now will prefer to use
the gobs of disk space and bandwidth that companies are falling all
over themselves to provide-- for free or cheap.
But hey, whatever works. As I write this, forth.org is down again.
John Passaniti
> Um... Why bother with either suggestion when there is RapidShare?http://www.rapidshare.com/
Um, because RapidShare sucks?
0) Files may be deleted after 90 days of inactivity.
1) You have to suffer with a URL like this: http://rapidshare.com/files/329583222/0102082110.jpg
2) There is no way to browse files. You would need a separate site or
listing for that.
3) When you do click on a link, you don't get the file-- you get a
screen where you have to first click to say if you're a "free user" or
a "premium user."
4) When you select "free user" you still don't get the file-- you get
a page with a download button.
5) When you finally click the download button, you do finally get the
file. Hooray!
So, tell me why RapidShare is the most awesome solution? A lack of
guarantees about file retention, clicky-clicky-clicky to download
anything, and constant advertising for themselves and their referral
scheme don't excite me. I'll admit that clicking on the video on
their site provides the cutest-ever German accents from the woman and
off-screen man that I've heard in a long time. But that hardly seems
like it will be much of a draw.
> Zip it. Upload it. Free storage. Free download. They provide a download
> link for it. Send the link to whomever... It's easiest file transfer
> option I've seen.
Your focus on delivery of the file misses the essential need: Finding
the content. At best, you could name the files like this:
http://rapidshare.com/files/329583222/issue-4-jan-1987.pdf
So if I knew that the article I wanted to read was in issue 4 or was
in January of 1985, I might know to download that file. But what if I
don't know? What if I only knew the article by author or title. Or
what if I didn't know author or title, but wanted to find all articles
about databases. How does RapidShare help with that?
Rod Pemberton
> On Jan 3, 2:02 am, "Rod Pemberton" <do_not_h...@nohavenot.cmm> wrote:
> > Um... Why bother with either suggestion when there is
RapidShare?http://www.rapidshare.com/
>
> Um, because RapidShare sucks?
>
> 0) Files may be deleted after 90 days of inactivity.
They still accept files without notifying you that they'll delete them after
90. I'm not sure whether they delete anyway... They didn't always limit
available time. If the limited time model doesn't work, they may convert
back to unlimited.
> 1) You have to suffer with a URL like this:
http://rapidshare.com/files/329583222/0102082110.jpg
You can always use:
http://tinyurl.com/
> 2) There is no way to browse files.
You've been given a direct link to the files you want. Why would you want
to browse through them?
> You would need a separate site or
> listing for that.
True.
> 3) When you do click on a link, you don't get the file-- you get a
> screen where you have to first click to say if you're a "free user" or
> a "premium user."
Three clicks? You're complaining about three clicks? Man that's almost the
holy grail... But, the holy grail was patented (US).
Have you ever dl'd a file from Simtel (modern html-ized), CNET Download,
MajorGeeks, etc.? 15 clicks minimum. Numerous retries. Pop-ups galore.
Ads everywhere. Even sites without the ads and pop-ups aren't spectacular.
Have you tried to locate and download files from SourceForge or even the
FSF? 8 clicks minimum. That's assuming you can even remotely find what
you're looking for at the FSF. If you don't know "ld" is in the "binutils"
package, just how do you find it?
> So, tell me why RapidShare is the most awesome solution?
1) free, simple, quick
2) no need for login account or email account to upload or remove
3) not limited to those who can afford their own fileserver
4) no excessive bandwidth harrasment issues from your ISP
5) no need to leave your computer running
6) no need to bypass your firewall's security to transmit files
etc. One could almost steal one of Ronco's catch phrases and use here.
> Your focus on delivery of the file misses the essential need:
> Finding the content.
You've been given a direct link to the files you want. Why would you need
to locate them?
> So if I knew that the article I wanted to read was in issue 4 or was
> in January of 1985, I might know to download that file. But what if I
> don't know? What if I only knew the article by author or title. Or
> what if I didn't know author or title, but wanted to find all articles
> about databases.
That's what Yahoo or Google or fileshare app's are for: indexing.
Rod Pemberton
Rod Pemberton
> Where I live, standard cable data rates are
> 10Mbps down and 384kbps up,
Rates vary heavily based on connection location (big vs. small city, far vs.
near, data center vs.private), time of day, and unknowns. Currently, peak
cable (one of the major US...) data rate here is 9200kbps down and 5500kbps
up. So, it's probably 10Mbps rated. Typically, 3/4 of that is sustainable.
A period of slowness will be 1/2 of that. The rates have improved here year
over year, especially upload which saw a large increase. I think this
corresponded with the digital movie on demand network upgrades. Due to
that, I'm not sure if the upload speed is real or just due to a quick,
local, intermediate storage. Originally, I think the broadband was 6Mbps
down and maybe around 384kbps up.
RP
Josh Grams
> "John Passaniti" <john.pa...@gmail.com> wrote in message
> news:acbe00ed-cba6-4668...@d20g2000yqh.googlegroups.com...
>> screen where you have to first click to say if you're a "free user" or
>> a "premium user."
>
> Three clicks? You're complaining about three clicks? Man that's almost the
> holy grail... But, the holy grail was patented (US).
>
> Have you ever dl'd a file from Simtel (modern html-ized), CNET Download,
> MajorGeeks, etc.? 15 clicks minimum. Numerous retries. Pop-ups galore.
> Ads everywhere. Even sites without the ads and pop-ups aren't spectacular.
> Have you tried to locate and download files from SourceForge or even the
> FSF? 8 clicks minimum. That's assuming you can even remotely find what
> you're looking for at the FSF. If you don't know "ld" is in the "binutils"
> package, just how do you find it?
Three unnecessary clicks is still totally ridiculous. The existence of
other download sites which suck much worse doesn't change that...
Especially when, as John pointed out, there are other free alternatives.
--Josh
Anton Ertl
>Aren't scanned Forth Dimensions issues already available on forth.org? This
>is the link I have.
>http://www.forth.org/fd/contents.html
>
>It seems they have from June/July 1978 (Issue 1, vol. 1) to May/August 1999
>(Issue 21, vol. 2) as .pdf scanned images (not OCR'd, not text searchable).
>Except for Issue #21, forth.org has six volumes per issue (bi-monthly?).
>Well, they're there - if or when you're able to connect to forth.org's
>intermittent website...
Volumes 1-6 and 21 and an index for Volumes 1-6 is also available on
<http://www.complang.tuwien.ac.at/forth/forth-dimensions/>. The index
on the address given above also seems to cover only volumes 1-6.
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2009: http://www.euroforth.org/ef09/
John Passaniti
> > 0) Files may be deleted after 90 days of inactivity.
>
> They still accept files without notifying you that they'll delete them after
> 90. I'm not sure whether they delete anyway... They didn't always limit
> available time. If the limited time model doesn't work, they may convert
> back to unlimited.
No wait-- here's a better idea. Since they may delete files after 90
days, why don't we just assign someone who's job it is to redundantly
download the file every 89 days! It's so simple! We just reset the
counter! See, all this time I thought the goal was serving up Forth
content, but now I understand the *true* goal is to assist
RapidShare!
> > 1) You have to suffer with a URL like this:
>
> http://rapidshare.com/files/329583222/0102082110.jpg
>
> You can always use:http://tinyurl.com/
Sure, but then you lose any connection between a naming convention for
the issues and the URL. So if you wanted to find a specific issue,
how do you construct the URL? Oh, and then you have another problem--
so what if we want to move the file off RapidShare and onto some other
equally dumb service. How do you update the TinyURL? Or let's say
you don't care about updating the TinyURL, and you just create a new
one. So now you have two (or more) TinyURL's for the same content
floating around, one (or more) that doesn't work, and one that does.
How exactly does that help address the issue of reliability?
> > 2) There is no way to browse files.
>
> You've been given a direct link to the files you want. Why would you want
> to browse through them?
Where exactly is this direct link coming from? If I want a specific
issue, or if I want a specific article, how exactly do I get this
"direct link?" Where is the index? It has to be hosted *somewhere*,
so why can't that same place be where you download the files?
Why would I want to browse through them? Are you kidding? Are you
seriously going to tell me that when forth.org is actually available
(which incidentally, it is as I write this), that you haven't skimmed
the index looking at interesting articles to read?
> > 3) When you do click on a link, you don't get the file-- you get a
> > screen where you have to first click to say if you're a "free user" or
> > a "premium user."
>
> Three clicks? You're complaining about three clicks? Man that's almost the
> holy grail... But, the holy grail was patented (US).
I fail to see why more than one click is needed. In your
justification for using RapidShare, you're going out of your way to
overlook every problem it presents and ignoring that there are both
free and cheap solutions that eliminate every problem *and* that
present the content in a way that is actually useful. Hell, with a
blog (or Wiki), you not only have the opportunity to present the
content, but allow the user community to build on that, adding
comments, discussions, posting code, etc.
> Have you ever dl'd a file from Simtel (modern html-ized), CNET Download,
> MajorGeeks, etc.? 15 clicks minimum. Numerous retries. Pop-ups galore.
> Ads everywhere. Even sites without the ads and pop-ups aren't spectacular.
> Have you tried to locate and download files from SourceForge or even the
> FSF? 8 clicks minimum. That's assuming you can even remotely find what
> you're looking for at the FSF. If you don't know "ld" is in the "binutils"
> package, just how do you find it?
So how exactly does this work for you? When forth.org is up and you
decide to click on one of the links to a Forth Dimensions issue, do
you click and then curse the screen, "damn it, I got the content with
a single click! Why on earth didn't I get two levels of confirmation
and endless advertising!"
> > So, tell me why RapidShare is the most awesome solution?
>
> 1) free, simple, quick
> 2) no need for login account or email account to upload or remove
> 3) not limited to those who can afford their own fileserver
> 4) no excessive bandwidth harrasment issues from your ISP
> 5) no need to leave your computer running
> 6) no need to bypass your firewall's security to transmit files
Gosh, I get the same, and it's even easier. I can drag and drop files
from my desktop to a folder that's connected to any FTP, SFTP, or WEB-
DAV-accessible site. Drag, drop, there, done. I use that manage
files on both free and non-free systems. I don't have to worry about
the files disappearing after 90 days, they don't have funky URLs, and
people downloading the files don't have to clicky-clicky-click.
RapidShare is the ideal solution for people who don't know what other
*easier* free-to-cheap options are out there.
> You've been given a direct link to the files you want. Why would you need
> to locate them?
Again, please demonstrate this. I'm looking for all articles
published in Forth Dimensions that had to do with metacompilation for
a 6809. Please provide me the "direct link."
> > So if I knew that the article I wanted to read was in issue 4 or was
> > in January of 1985, I might know to download that file. But what if I
> > don't know? What if I only knew the article by author or title. Or
> > what if I didn't know author or title, but wanted to find all articles
> > about databases.
>
> That's what Yahoo or Google or fileshare app's are for: indexing.
Google only indexes what it can see. We've already established that
RapidShare provides no way to browse files, so Google will be unaware
of any files stored there. At best, someone would need to post a
separate newsgroup message with an index, or post the same to a web
site. And how exactly do "fileshare apps" solve any problem? What's
you're solution-- put all the issues on BitTorrent?
So let's summarize. I provided three different solutions that range
from cheap to free. The solution of using a free blog (such as on
WordPress or Blogger) provides not only the means to provide the means
to search for content (full text, tags), but also gives people the
means to comment on issues, provide updates, cross-references, etc.
Don't like blog structure? There are free Wiki's that would allow not
just access to the files, but also allow people to create entire pages
for comment, discussion, posting code, or inventing their own index.
And it's *free* and easily movable elsewhere if and when that was
needed. Oh, did I mention, free?
You're right that the person publishing the files would have to log
in. Wow, that's a really compelling argument. Oh wait-- if
structured as a public Wiki, you wouldn't even need to do that.
So what's the real deal here? You looking for referral credits on
RapidShare?
MarkWills
Tracing route to taygeta-fe1-0.2.core-01.mbay.net [206.55.246.130]
over a maximum of 30 hops:
1 63 ms 99 ms 99 ms BThomehub.home [192.168.1.254]
2 53 ms 53 ms 57 ms 217.47.95.122
3 51 ms 50 ms 51 ms 217.47.95.193
4 52 ms 48 ms 48 ms 217.47.105.10
5 49 ms 47 ms 49 ms 217.41.174.29
6 54 ms 54 ms 54 ms 217.41.218.246
7 56 ms 54 ms 52 ms 217.41.218.54
8 55 ms 55 ms 52 ms 217.47.186.242
9 53 ms 54 ms 58 ms core1-pos15-3.edinburgh.ukcore.bt.net
[217.32.169.217]
10 68 ms 62 ms 61 ms core1-posch1.birmingham.ukcore.bt.net
[62.6.201.77]
11 69 ms 70 ms 73 ms core1-pos0-1-4-0.ilford.ukcore.bt.net
[62.6.204.10]
12 66 ms 67 ms 67 ms
core3te-0-3-1-0.telehouse.ukcore.bt.net [62.172.102.9]
13 68 ms 66 ms 65 ms ge-2-1-0.mpr1.lhr2.uk.above.net
[195.66.224.76]
14 138 ms 139 ms 135 ms so-1-1-0.mpr1.dca2.us.above.net
[64.125.31.186]
15 142 ms 149 ms 144 ms so-1-2-0.mpr1.lga5.us.above.net
[64.125.26.102]
16 227 ms 223 ms 226 ms so-2-1-0.mpr1.sjc2.above.net
[64.125.26.229]
17 226 ms 241 ms 245 ms 64.124.193.45.available.above.net
[64.124.193.45]
18 763 ms 245 ms 249 ms ge-0-0-0-742.cr2.lma4.got.net
[207.111.207.44]
19 229 ms 226 ms 224 ms ge-0-0-0-741.cr1.scz1.got.net
[207.111.207.9]
20 225 ms 225 ms 222 ms ge-1-0-10.er1.scz1.got.net
[207.111.198.51]
21 226 ms 222 ms 224 ms mbay.got.net [207.111.222.6]
22 * 253 ms 225 ms taygeta-fe1-0.2.core-01.mbay.net
[206.55.246.130]
Trace complete.
I was surprised at the number of hops, 17 of which appear to be in the
UK!
It's just a hunch, but I think Forth.org (or rather, the Taygeta
server) is on a domestic home connection and is using dynamic ip. The
home connection would account for the poor bandwidth (I have a 10 mbit
connection, but cant get more than 30k a second from it) and the ip
address changing would account for the site being up and down like a
yo-yo. Again, just theorising, but perhaps the 'down time' is the
delay in DNS syncning after the ip address change. That or the server
keeps crashing!
It's just a theory. I'll check the ip address next time it goes down
and see if it changes. The site has been down three times today that I
know of.
Ed
It appears that many (all?) of the FD scans on www.forth.org now include
OCR'd text under the images. Originally they didn't.
John Passaniti
> It's just a hunch, but I think Forth.org (or rather, the Taygeta
> server) is on a domestic home connection and is using dynamic ip.
Maybe, but I'd guess not. The site is down as I type this, but the
last captured page on archive.org gives as an address for the Taygeta
office that when viewed with Google Maps, shows a large office
building. That, combined with the name "taygeta-
fe1-0.2.core-01.mbay.net" suggests a dedicated connection of some
sort, so it probably isn't dynamic. My guess is simply that the
server itself is down.
> The
> home connection would account for the poor bandwidth (I have a 10 mbit
> connection, but cant get more than 30k a second from it) and the ip
> address changing would account for the site being up and down like a
> yo-yo.
mbay.net offers switched 56k and ISDN-- both typically provided
guaranteed service levels, but are slow. mbay.net also resells DSL
connections, which are typically asymmetric and sluggish.
Rod Pemberton
> When forth.org is up and you
> decide to click on one of the links to a Forth Dimensions issue, do
> you click and then curse the screen, "damn it, I got the content with
> a single click! Why on earth didn't I get two levels of confirmation
> and endless advertising!"
Funny!
No, I got seriously PO'd that it took 121 clicks to download them all -
instead of a single click to download them all at once. Where does your
single click come in again?
It seems you want an "all-in-one" model or solution. An "all-in-one"
model - specifically for academic papers - is CiteSeer (or CiteSeerX). It
archives content, converts content to standard formats, indexes content,
searches content, and generates links directly to their archived content.
> RapidShare is the ideal solution for people who don't know what other
> *easier* free-to-cheap options are out there.
Oh, I seriously disagree. I've been using the "Internet" since the late
'80's from 7-bit telnet, rz/sz, "anonymous" ftp sites with index list,
Kermit, KA9Q, zmodem, upload/download ratios, gopher, to modern html
webpages with Google/Yahoo, and various P2P fileshares, etc. The only
"better" solution, IMO, is directly from a webpage which was found via a
search (Yahoo, Google, Google Groups) for the specific info you wanted.
CiteSeer isn't bad if you want academic content. eDonkey (or the clone
Emule) wasn't too bad. It indexed. It had minimal firewall bypass. But,
it had problems completing files, it had in-network connection issues, and
minimally adhered to a variations of the old ratios concept: upload speed
vs. download speed and shared quantity vs. download quantity. Upload and
download ratios both built and killed off dialup BBSes - due to high, then
low costs of telecommunications. Unfortunately, the last time I looked
BitTorrent - known for large files and/or decent speed - didn't index squat.
WebArchive doesn't index either. You need to know the ancient link.
> > You've been given a direct link to the files you want. Why would you
need
> > to locate them?
>
> Again, please demonstrate this. I'm looking for all articles
> published in Forth Dimensions that had to do with metacompilation for
> a 6809. Please provide me the "direct link."
You misunderstand. If I were to upload them to RapidShare, they'll give me
a link(s). If I posted the link here, you'd get a link for .zip or bzip2 -
with all of them in one file because that's the way I'd package them. You
don't need to locate them. They are all there in a single file at a single
link. You may need to locate the articles within. But, indexing is
different from locating them.
> structured as a public Wiki
Wiki's record your IP. IMO, that's not a problem. But, they also *display*
your IP to the *entire* reading public. IMO, that's a serious problem.
Haven't you read about denial of service (DoS) attacks against "anonymous"
users whose IP's were displayed? Wikipedia, the most famous Wiki,
"regulars" blame "anonymous" users for everything wrong with Wikipedia. The
real problem is Wikipedia's ideology is faulty. Cliche's like "too many
cooks spoil the broth" and "familiarity breeds contempt" accurately describe
what happens when you allow equality among a large set of people of diverse
intellect. You have to remember that intelligence, experience, and
knowledge aren't evenly distributed. I.e., most in Wikipedia's large "room"
of contributors are average or worse. One primary result of Wikipedia's
ideology is that there is no control over the removal of content on
Wikipedia. It should take three people to remove content, but only one to
add. This would prevent Wikipedia needing the help of socially aware Univ.
Prof's from having to rewrite advanced topics every six to eight months
because the articles are dumbed down to the level of a fifth grader over
time. They have no article rating system: general, in-depth, advanced,
academic, genious - that could be used to control addition and removal of
content and separate contributors by groups.
> So what's the real deal here?
Your solutions weren't as easy, quick, anonymous, free as other solutions,
IMO. Your solutions may index articles, but if you can download all the
magazines with one click - then you can peruse them at your leisure. If no
index is available, you can construct your own index as I'm doing.
Rod Pemberton
PS You can keep going if you want, but I'm winding down my responses.
Albert van der Horst
John Passaniti <john.pa...@gmail.com> wrote:
>On Jan 3, 2:43=A0pm, MarkWills <markrobertwi...@yahoo.co.uk> wrote:
>> It's just a hunch, but I think Forth.org (or rather, the Taygeta
>> server) is on a domestic home connection and is using dynamic ip.
>
>Maybe, but I'd guess not. The site is down as I type this, but the
>last captured page on archive.org gives as an address for the Taygeta
>office that when viewed with Google Maps, shows a large office
>building. That, combined with the name "taygeta-
>fe1-0.2.core-01.mbay.net" suggests a dedicated connection of some
>sort, so it probably isn't dynamic. My guess is simply that the
>server itself is down.
I couldn't care less if the server is down for 80% of the time.
I couldn't care less if in a library the seventh volume of Euclid's
complete works is out 80% of the time, as long as I can reserve it.
Stable references (URL's) are far more important than uptime of
a server, but I must admit, that I also draw the conclusion that
I site no longer exists as soon as I can't access it.
<SNIP>
>mbay.net offers switched 56k and ISDN-- both typically provided
>guaranteed service levels, but are slow. mbay.net also resells DSL
>connections, which are typically asymmetric and sluggish.
>
I host archive material on my home server. Also asymmetric
and somewhat slow. Who cares? It is convenient.
Groetjes Albert
--
--
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
idknow
Why not just put the files on Sourceforge?
They don't delete files. the only issue would be who will be primary
contact? The one who sets up the project most likely.
David N. Williams
> [...]
>
> It appears that many (all?) of the FD scans on www.forth.org now include
> OCR'd text under the images. Originally they didn't.
From http://www.forth.org/fd/contents.html
A note on these Forth Dimensions files:
"I have processed these Forth Dimensions PDF files thru the
"optical character reader" (OCR) and "reduce file size"
functions of Adobe's "Acrobat Version 7". This does a pretty
good, but not perfect, job of recognizing continuous
English-language text characters, but I have not proof-read
any of the text, so there will certainly be errors."
But I don't see the original scans anymore, and in view of the
above statement about proofing, they need to be available as
well, no?
-- David
Albert van der Horst
Rod Pemberton <do_no...@nohavenot.cmm> wrote:
<SNIP>
>
>> structured as a public Wiki
>
>Wiki's record your IP. IMO, that's not a problem. But, they also *display*
>your IP to the *entire* reading public. IMO, that's a serious problem.
>Haven't you read about denial of service (DoS) attacks against "anonymous"
>users whose IP's were displayed? Wikipedia, the most famous Wiki,
>"regulars" blame "anonymous" users for everything wrong with Wikipedia. The
>real problem is Wikipedia's ideology is faulty. Cliche's like "too many
>cooks spoil the broth" and "familiarity breeds contempt" accurately describe
>what happens when you allow equality among a large set of people of diverse
>intellect. You have to remember that intelligence, experience, and
>knowledge aren't evenly distributed. I.e., most in Wikipedia's large "room"
>of contributors are average or worse. One primary result of Wikipedia's
>ideology is that there is no control over the removal of content on
>Wikipedia. It should take three people to remove content, but only one to
>add. This would prevent Wikipedia needing the help of socially aware Univ.
>Prof's from having to rewrite advanced topics every six to eight months
>because the articles are dumbed down to the level of a fifth grader over
>time. They have no article rating system: general, in-depth, advanced,
>academic, genious - that could be used to control addition and removal of
>content and separate contributors by groups.
These disadvantages were known from before it started and
led many to believe that no good good come out of the concept.
However apparently, almost all "fifth grader" can recognize and stand
in awe of contributions of "college professors", such that they
have to rewrite their content *merely* once in a while.
So Wiki works out to the good of mankind, and I don't mind that they
can see my IP-address.
I hope you don't succeed into discouraging people from contributing
to Wiki.
You don't drive a car, do you? The fear for an accident must be
much worse than the fear for a DOS attic on your computer, because
the consequence are so much more severe?
>
>Rod Pemberton
>PS You can keep going if you want, but I'm winding down my responses.
>
Groetjes Albert
David N. Williams
Hummn... Reading further:
"So with such text you can use Acrobat or the free Acrobat
Reader to view these files as an image, and to search for
words, or to copy words, lines, or blocks of text, which are
"text behind the image". But for any critical application, you
must proof read the resultant text."
Which seems to imply that the files still contain the scanned
images?
-- David
John Passaniti
> No, I got seriously PO'd that it took 121 clicks to download them all -
> instead of a single click to download them all at once. Where does your
> single click come in again?
Gee, when hosted on a system one actually has control over, one is
free to offer the content in any form desired. Even the complete
collection in a single click, which is a new requirement you pulled
out of thin air. Even better, that same content can be served using
protocols like rsync, allowing for efficient mirroring and adding
further redundancy.
Oh, and incidentally, you didn't need to click 121 times. You could
use a tool such as wget to grab all the files in a single command.
Here, allow me to demonstrate:
wget -m -np http://forth.org/fd/FDcover.html
Ta-da! I hit enter, went downstairs to make myself a nice sandwich,
and now I have a copy of that entire section of the web site.
Granted, I missed out on practicing my hand/eye coordination as you
did, but my hands were free to stuff my face with delicious deli meats
and cheeses.
Oh, and that epic download-everything-single-click requirement you
invented for some reason? Gee, that sounds hard:
tar -cjf fd-site.tar.bz2 fd
But a lot of good that does you. If only there was a way to put that
online so that people could download it as a single click. Why, here
you go! 906 megabytes of goodness:
http://japanisshinto.com/stuff/fd-site.tar.bz2
Enjoy. Except now you'll probably bitch about not being able to get
to the index and download them individually. Good point! Here you
go, a complete (if temporary) mirror of the Forth Dimensions part of
the site:
Wow, that took me all of 15 minutes to download and setup. You were
right-- RapidShare is clearly the way to go. Setting up a web site is
so *hard*!
> It seems you want an "all-in-one" model or solution.
Ummm, no. My primary goal is a reliable repository for Forth
Dimensions, and I've now given five different ways to do that for free
or cheap *and* set up a demonstration mirror. A secondary goal would
be to not just have a dumb repository, but something that exploits the
fact that we're living in 2010 and we have a host of tools that can be
used to turn those documents into conversations. I'll probably do
that next, since apparently the concept is so difficult to grasp.
> > RapidShare is the ideal solution for people who don't know what other
> > *easier* free-to-cheap options are out there.
>
> Oh, I seriously disagree. I've been using the "Internet" since the late
> '80's from 7-bit telnet, rz/sz, "anonymous" ftp sites with index list,
> Kermit, KA9Q, zmodem, upload/download ratios, gopher, to modern html
> webpages with Google/Yahoo, and various P2P fileshares, etc.
But apparently you haven't been using the Internet long enough to know
that you didn't need 121 clicks to get the content you wanted using a
common utility that was first released around 14 years ago.
> The only
> "better" solution, IMO, is directly from a webpage which was found via a
> search (Yahoo, Google, Google Groups) for the specific info you wanted.
Which is where I started. I'm glad you finally see the light.
> CiteSeer isn't bad if you want academic content. eDonkey (or the clone
> Emule) wasn't too bad. It indexed. It had minimal firewall bypass. But,
> it had problems completing files, it had in-network connection issues, and
> minimally adhered to a variations of the old ratios concept: upload speed
> vs. download speed and shared quantity vs. download quantity. Upload and
> download ratios both built and killed off dialup BBSes - due to high, then
> low costs of telecommunications. Unfortunately, the last time I looked
> BitTorrent - known for large files and/or decent speed - didn't index squat.
> WebArchive doesn't index either. You need to know the ancient link.
Fascinating. In the time you went down this nostalgic trip down
memory lane with a side trip down dial-up BBS-land, I set up a mirror
of the site.
> You misunderstand. If I were to upload them to RapidShare, they'll give me
> a link(s). If I posted the link here, you'd get a link for .zip or bzip2 -
> with all of them in one file because that's the way I'd package them. You
> don't need to locate them. They are all there in a single file at a single
> link. You may need to locate the articles within. But, indexing is
> different from locating them.
Now I understand. And now, I don't care. Aside from replacing the
ridiculous RapidShare with a file that you or anyone else is free to
download from my site, I'm next moving on to making this content
social.
> > structured as a public Wiki
>
> Wiki's record your IP.
Most every web site records your IP address. Shocking!
> IMO, that's not a problem. But, they also *display*
> your IP to the *entire* reading public. IMO, that's a
> serious problem.
There are currently about 23-gazillion different Wiki's out there in
the world. Where you got the idea that *all* of them necessarily
expose your IP address is beyond me. But let's say for the moment
that your statement was true, and that all Wiki's desperately want to
show your IP address to evil hackers. Now sit down because this is
going to be a shock, but SINCE THE SOURCE CODE IS AVAILABLE, YOU CAN
CHANGE IT. Wow!
> Rod Pemberton
> PS You can keep going if you want, but I'm winding down my responses.
Nah, consider this my last response to you on this. I'm probably
going to set something more permanent and social this week. You're
free to ignore it, and offer the world that awesome RapidShare "direct
link." Have fun.
Ed
Don't they? Viewing a copy of FD-V20N2.pdf (3.7Mb) I d/loaded
some time ago, I see the pages as a graphic images. Doing a text
search finds text.
Some PDF viewers (e.g. Foxit Reader) allow one to switch between
the images and the underlying OCR'd text. OCR quality of the FD
text isn't fantastic but it's better than nothing.
When the first several volumes of FD were originally scanned,
they only included images. This may still be the case i.e. only
the later added volumes include the OCR'd text.
David N. Williams
> David N. Williams wrote:
.. [...]
>>
>> Hummn... Reading further:
>>
>> "So with such text you can use Acrobat or the free Acrobat
>> Reader to view these files as an image, and to search for
>> words, or to copy words, lines, or blocks of text, which are
>> "text behind the image". But for any critical application, you
>> must proof read the resultant text."
>>
>> Which seems to imply that the files still contain the scanned
>> images?
>
> Don't they? Viewing a copy of FD-V20N2.pdf (3.7Mb) I d/loaded
> some time ago, I see the pages as a graphic images. Doing a text
> search finds text.
Seems to be the case. I just wasn't aware that .pdf was so
capable, which hindered my processing the implication of the
quote above.
> Some PDF viewers (e.g. Foxit Reader) allow one to switch between
> the images and the underlying OCR'd text. OCR quality of the FD
> text isn't fantastic but it's better than nothing.
>
> When the first several volumes of FD were originally scanned,
> they only included images. This may still be the case i.e. only
> the later added volumes include the OCR'd text.
AFAICT, the earlier ones are searchable as well, and the note at
the site seems to say they've all been processed.
The compression seems pretty good, too. At least I didn't win
bytes by tar'ing and zip'ing the wbole thing.
-- David
Anton Ertl
>Wiki's record your IP. IMO, that's not a problem. But, they also *display*
>your IP to the *entire* reading public.
Wikipedia does not for contribution by registered users; it displays
the user name instead.
>IMO, that's a serious problem.
>Haven't you read about denial of service (DoS) attacks against "anonymous"
>users whose IP's were displayed?
No, and I have not experienced that, even though I did some edits
where my IP was displayed.
Ed
Don't know what use others are making of FD i.e. whether it's
just nostalgic reading or something more serious. If it's the
latter, it may worth extracting the technical articles/letters
and presenting them in a more user-friendly form e.g. html.
On occasion I've had to scan/OCR/proofread code listings
from printed text, I found it wasn't much more effort to html
the entire article. If each person did that for the articles in
FD that interested them, a useful compilation could quickly
be built. Being html, the separate articles would be easy to
collate and index.
The only practical difficulty I see is the issue of copyright.
Early FD was entirely in the public domain (a smart move).
Later, code from articles were p/d but not the text. In some
instances, even code listings had restricted use.
MarkWills
from printed text, I found it wasn't much more effort to html
the entire article."
Hi Ed,
What tools did you use to produce a HTML version of the OCRd text?
Regards
Mark
Ed
Any reasonable OCR program (Omnipage, FineReader etc) will have
options to set the output format/style/doc-type including RTF, Word,
HTML).
For myself, I find it easier to OCR to plain ascii text to remove all
formatting other than paragraphs and do a HTML from that. The
HTML version has minimal formatting - Times Roman font for the
article text, and Courier for program listings and any Forth words
in the text body. If the original had bold/italics/super/subscript/math
formulae etc, those are duplicated.