Try using a regular newsreader. They work
far better than the miriad of web-based services.
A.newsrc is very useful. It keeps a record of which articles you have read, so then when you come back a day (or an hour) later, only new articles are presented.
Okay, so "alt.binaries.porn.children"
contains child porn?
How about these groups?
alt.binaries.porn.not-children
alt.binaries.porn.ages.21-71
alt.binaries.porn.adults.honest
Obvious attempts at reverse psychology or
correctly named groups?
Perhaps ISPs should lose all large base64/uucode
articles and tell people that usenet is for discussion, and
that large files should be distributed of web sites.
I read about 30 groups without filtering,
all of which are roughly 1% junk.
(Compare that with slashdot at -1.)
A big problem are the binaries. Usenet was
designed for discussion, not for transport of
large files. Perhaps ISPs should filter out the
all binary articles (porn or not), and tell them to
put up websites if they want to distribute large files.
Bill, find . -type f | xargs grep "Content-Transfer-Encoding: base64"
Usenet was designed as a discussion medium, not a means to distribute
large files.
It works great at the job it was designed (even better than
slashdot, IMHO)
At the job of distributing large files... Each ISP keeps a copy,
regardless if anyone wants it or even asked for it. That is not a good
utilisation of resources.
Bill, has a newsreader which only presents new articles.
Re:Yet another knee-jerk, anti-WIPO /. Posting
on
Is It OK To Sucks?
·
· Score: 1
Next week, why people who don't lock thier front doors deserve to
have thier property taken away.
I have my foreground colour set to the colour I like.
I have by background colour set to the colour I like.
I have set the typefaces to ones I like the look of.
A web designer attempting to force thier own
preferences on me will either produce a web site
of equal quality, or lower quality, than if you just
left it alone.
If you don't waste time fussing about how it will look, you can instead work on the interesting stuff.
That will produce a higher quality web site.
Bill, has slashdot in "light mode" because it works better in Netscape.
One way to check the security of completed blocks is to have
each block also calculate some incidental security check.
For example, as well as making note of all keys that decrypt
to "The secret word is", it could also make note of all keys that
decrypt to "The". In each block of 2^32 keys, there should be around
256 which decrpyt to this.
If a returned block contains (say) 200 or more keys which
decrpyt to "The", these can all be checked by repeating the
decrpytion of only these 200-or-more keys (which should take
milliseconds). If any do not decrypt
to "The", then the returned block is definately bogus.
If all the 200-or-more match, then it's reasonable that the
client did it's job properly, since there's no way (in theory)
to work out which of the 2^32 keys would result in this decryption,
without going through them all one by one, which is exactly
the job the client is being paid for.
The returned block contains below 200 keys, all of which decrpyt to
"The" then this returned block may be bogus. It's possible that
this key block is a freak where only a few keys decrypt this way,
but it's more likely that the client stopped short and retrurned
the incomplete block, hacked to look like a completed block.
As an extra security check, the central server could re-issue a
proportion (say 10%) of completed blocks again, to see if other clients
give the same result. If two clients return different completed blocks
for the same job, then one of them is bogus. You can run the
block yourself to find out who.
Re-issuing will result in a slow-down, so is best used sparingly.
Slashdot Mirror
In days of old, hardware makers sold thier hardware at a profit, and pretty much left the software developers to do thier own thing.
Games were written by people in thier own homes. No need for big software houses.
Technology has improved, but the games have not. The innovative people can't afford thier own copy of the SDK.
The sooner the world gets away from the con of software sales subsidising loss-making hardware, the better.
Bill, jet set.
How I long for the days of the Sinclair Spectrum.
Sinclair sold his hardware at a profit, and took no cut of the software.
All the best games were written by teenagers in thier bedrooms, and they could compete fairly with the big SW publishers.
That's a business model I can respect.
In the years since, the technology has got better, but the games have got worse. The current crop has no soul.
Bill, de do do do do do.
Give them a pre-compield binary executable and specify that the contract is for that only.
Repeat after me, "There is more to usenet than the alt.* hierarchy."
Bill, "Slashdot is just a load of links to some goatsex website."
Try using a regular newsreader. They work far better than the miriad of web-based services.
A .newsrc is very useful. It keeps a record of which articles you have read, so then when you come back a day (or an hour) later, only new articles are presented.
Why oh why can't slashdot do this?
Bill, slashdot: 1-59,61-97
Usenet is full of junk? Try reading slashdot at -1 for a while.
Bill, uses a newsreader.
Somewhere in the fifth dimension, a copy of you will have just won.
Bill, wondering how long it would take to phone all these copies.
How would you feel about someone who hangs around all day with a notebook, taking notes.
Bill, just don't point them in my place, okay.
I (and a _lot_ of other people) would much rather put up with tacky and inconvenient ads than having to pay a fee.
Alas, tacky and inconvienient ads are becoming less and less able to pay thier way.
Deja carried adverts, and despite being a popular service (look at the reaction to deja suddenly getting switched off) it made a loss.
Bill, just take the money.
So how much would you pay for a years access to the archives?
I would happily hand over $10 a year. I really do miss dejanews in it's heyday.
Come to think of it, I would pay for access to Google web search as well, so long as the fluff is removed.
Bill, a liker of usenet.
Okay, so "alt.binaries.porn.children" contains child porn?
How about these groups?
Obvious attempts at reverse psychology or correctly named groups?
Perhaps ISPs should lose all large base64/uucode articles and tell people that usenet is for discussion, and that large files should be distributed of web sites.
Bill, loves Usenet, and hates to see it abused.
Usenet is not dead (yet).
I read about 30 groups without filtering, all of which are roughly 1% junk. (Compare that with slashdot at -1.)
A big problem are the binaries. Usenet was designed for discussion, not for transport of large files. Perhaps ISPs should filter out the all binary articles (porn or not), and tell them to put up websites if they want to distribute large files.
Bill, find . -type f | xargs grep "Content-Transfer-Encoding: base64"
Usenet was designed as a discussion medium, not a means to distribute large files.
It works great at the job it was designed (even better than slashdot, IMHO)
At the job of distributing large files... Each ISP keeps a copy, regardless if anyone wants it or even asked for it. That is not a good utilisation of resources.
Bill, has a newsreader which only presents new articles.
Next week, why people who don't lock thier front doors deserve to have thier property taken away.
Bill, president of Billville.
It's Media3 that are getting blacklisted, not Peacefire. If Peacefire left, the RBL listing would not follow them.
The boycott is to put pressure on Media3 to change thier policy. If Peacefire leave, M3 get less money, and so increases to the pressure.
and has it been fixed?
With the exception of a few moderated groups, and some alt groups that rabidly protect their turf from spammers, Usenet is a wasteland of spam.
Am I reading the same Usenet? I read about 20 groups, and the spam problem is very low these days.
comp.lang.c gets about 250 articles a day, and about one or two of those are junk. (Main problem these days is "top posting")
Bill, munching some CLC rock.
The RBL has put pressure on Peacefire (and others) to consider switching from Media3 to another hoster which does not host those who sell Spamware.
If Peacefire does move, Media3 will lose money. It's an incentive for Media3 to change thier rules to stop Spamware from thier hosts.
If you don't like the RBL, don't use it.
Bill, doesn't use any MAPS services, but (morally) supports it's work.
I read that as "Hey, these computer things are going to get really powerful."
That's a long way from the precision of "x2 every 18 months."
Bill, an Englander looking in.
Find yourself a load of decaying atoms. Pick one atom and look at it for it's half life.
If it decays, write down a "1". If it doesn't write down a "0".
Do the same for 128 atoms. At the end, there are 2**128 copies of yourself, each with a different 128 bit number.
Apply this key to your cypher-text. If it works, tell the other (2**128)-1 copies of yourself what the key is.
That last part is the tricky bit. See if you can use quantum interference.
Are they finally going to stop replacing tried and tested protocols and data formats with thier own cheesy alternatives?
My original title was "How about EnCCS?"
Not my fault guv!
I have my foreground colour set to the colour I like. I have by background colour set to the colour I like. I have set the typefaces to ones I like the look of.
A web designer attempting to force thier own preferences on me will either produce a web site of equal quality, or lower quality, than if you just left it alone.
If you don't waste time fussing about how it will look, you can instead work on the interesting stuff. That will produce a higher quality web site.
Bill, has slashdot in "light mode" because it works better in Netscape.
Using the RC5-64 challenge as an example...
One way to check the security of completed blocks is to have each block also calculate some incidental security check.
For example, as well as making note of all keys that decrypt to "The secret word is", it could also make note of all keys that decrypt to "The". In each block of 2^32 keys, there should be around 256 which decrpyt to this.
If a returned block contains (say) 200 or more keys which decrpyt to "The", these can all be checked by repeating the decrpytion of only these 200-or-more keys (which should take milliseconds). If any do not decrypt to "The", then the returned block is definately bogus.
If all the 200-or-more match, then it's reasonable that the client did it's job properly, since there's no way (in theory) to work out which of the 2^32 keys would result in this decryption, without going through them all one by one, which is exactly the job the client is being paid for.
The returned block contains below 200 keys, all of which decrpyt to "The" then this returned block may be bogus. It's possible that this key block is a freak where only a few keys decrypt this way, but it's more likely that the client stopped short and retrurned the incomplete block, hacked to look like a completed block.
As an extra security check, the central server could re-issue a proportion (say 10%) of completed blocks again, to see if other clients give the same result. If two clients return different completed blocks for the same job, then one of them is bogus. You can run the block yourself to find out who.
Re-issuing will result in a slow-down, so is best used sparingly.