In trying to explain open source software to friends and family, I often run into the "you get what you pay for" line. My standard answer now goes something like this...
Open source contrasts heavily with proprietary software, where it is in fact true that you get what you pay for.
With open source software, more often than not, you get an awful lot more than you paid for. If by some fluke you actually do get what you paid for, you really have no grounds for complaint......
All fine and good, except that settlements do not establish precedent. Court _decisions_ do. In any future case, the court will look for similiar case law. If BT is really attempting to establish a precedent that they can use to go after bigger fish, the will _not_ accept a settlement offer, no matter how small or big.
I've got an aiwa cdc-mp3, works great for me. I haven't listened to the radio during my commute in literally months. and it _really_ helps during long road trips; a cd holds a _lot_ of high-bitrate music.
the only minor annoyance is that it won't restart mid-track. for me, not such a big deal.
- mark
Re:Tracking encrypted communications
on
A New Kind of War
·
· Score: 1
Cryptonomicon had a fairly good example of this. it's fictional, but the concepts apply.
as the story unfolded, the Japanese had a code that we hadn't broken. but could track when such encoded messages were sent, and what traffic they spurred. so a single message spawns many other messages, possibly using codes we have broken. in the particular instance in the story, the resulting messages went out ordering all available Japanese excavation and mining engineers to the Phillipines. hmmm... the Japanese are digging something BIG in the Phillipines. and it is important enough to them that the original orders were encrypted with this new code......
so, without decoding the original message, you can at least analyze the ripples it produces, and learn _something_ about the contents.
this is pretty much what happened to informix. they bought illustra and rolled illustra ordmbs tech into informix to create informix 9. they did an amazing job prepping their sales force for the release. so when the release slipped, they'd already convinced their users that version 9 was _so_ much better than version 7. naturally sales of the informix 7 slipped.
the story ends with a buyout by ibm. pretty much for market share, since db2 had most of the same features as informix 9.
so, customers do actually hold out for "so much better" product you've primed them for. and it certainly doesn't help your bottom line in the short term.
is that hits or pages for cnn? I count about 60 images on the cnn.com front page. still works out to an impressive ~850 _pages_ for 50k _hits_, but not as impressive as 50k pages would have been... that's the whole point of the comment earlier about cnn dropping their contract with akamai. with akamai, cnn would have been serving those 850 pages, but never would have seen the 50,000 images requests...
"first research laboratory started by a software company"
Xerox isn't a software company, at least not primarily. of course, one could argue that Microsoft is no longer exclusively a software company (keyboards, mice, xbox...), but that's another discussion. they were certainly only a software company when the founded Microsoft Research, and I believe this is the point the poster was attempting to make...
if I remember correctly (it's been a few years since I heard him describe the work), they worked out a closed form solution to the equations that govern fluid flow over a transonic body. given that, it isn't too hard to engineer a fuselage that doesn't generate a boom as it reaches mach 1. oddly enough, it looks suspiciously like a missile.
actually, it can be a bit simpler than that... if you use qmail as the mail server for your domain, then everything up to '-' is evaluated as an address for delivery. I routinely give
from-[NameOfCompany]@[mydomain]
as my email address when I'm forced to register. all such mail is then delivered to from@[mydomain] by qmail. the advantage here is you aren't creating a million aliases to keep track of; and it is _very_ easy to see who sold my name to whom.
3. MySQL's "text indexing" is useless. The evaluation function returns every record that contains any of the search terms; there is no way I've found to require all search terms. No documentation, of course.
actually, in the
mysql fulltext search docs
the mysql guys do a decent job of detailing the _two_ options for querying the fulltext index. putting the "match (a,b) against ('words')" in the "select" list returns the relevance rank against all rows in the table, including those with zero relevance. if you put the match clause as part of the "where" clause then mysql returns only those rows with non-zero relevance, ordered by the relevance score (but the relevance score is not one of the return columns).
- mark
Re:Roll your own archive...
on
Deja For Sale
·
· Score: 1
actually, since 3.23.23, mysql supports fulltext indexing of varchar and text columns in myisam tables. the results are naturally much faster than a like query against a column or columns.
indexing your archive might be a good test of their text index scalability...
actually, the akamaized link should stay the same (e.g. a935.g.akamaitech.net/...). akamai's dynamic dns chooses the best box to serve it to you, so your squid cache should only contain one copy.
at first I thought this sounded like the greatest feature ever: an ad-blocker built into the browser. no more external filtering proxies... yeah!
but then I started thinking about me as a developer and sites that I've built to make heavy use of services like akamai. now _none_ of my images will show up in your browser because all of the images are served out of akamaitech.net (akamai's distributed network). now it doesn't sound so great...
unless I could setup akamaitech.net as an exception, and therefore acceptable to fetch images from (but how many average joe users of netscape would actually understand why my page looked bad in the first place?). but then, if I allow akamai images, the ad banner folks will just start akamaizing their delivery (not a bad idea to begin with), and I'm back to square one...
you probably want to read this article on the real complexity involved in this sort of localization, and the pitfalls inherent in the sort of substitution you suggest.
suffice it to say, it's a nightmare for anything beyond nice and friendly iso8859-1. the author uses an amusing anecdote about a simple localization of a simple feedback message into Chinese, Arabic, Russian and Italian...
- mark
and if closest family is 1000 miles away???
on
LonelyNet
·
· Score: 1
in that case email and talk are about the only way to have any meaningful and frequent interaction with family. bet they didn't take that into account, huh?
as I see it, the right to pursue gainful employment is always going to win out in the end.
yes, I had to sign a non-compete when I started with my current employer, but I've never worried about it. for one thing I live and work in california where (as previously mentioned) these sort of contract provisions have no teeth.
beyond that, it all comes down to how they attempt to define a "competitor". they might conceivably be able to stop you from going to work for their number one rival (and even that is a long shot here in california), but there is absolutely no way they can claim the primary business to be "web" and bar you from working for _any_ other web firm. ludicrous!
it really depends on the branch of mathematics you are interested in (of course)... I'm personally into Analysis, so I was looking at (and continue to look at)
University of British Columbia Johns Hopkins Columbia
though keep in mind that there are other schools that do Analysis well, these are just the top schools on my radar.
also note: last time I checked, the market for a PhD in Mathematics was abyssmal. my senior year of undergrad I was on the search committee for _one_ tenure-track position at a small (but good) school. we had to wade through ~1000 serious applicants, and I've heard that other job postings are similiarly competitive. this is why I am still in private industry...
the issue isn't just the inexpensive board, it is the cost and performance of the chip (let's _not_ start any g3 vs. piii religious wars here). and for me personally the power consumption of the chip is important as well... it might be easier to run a linux-intel box as my network gateway, but I like my netwinder because of the _way_ low power consumption. likewise the ppc chips perform much better than intel hardware in power consumption arena (ever used a g3 laptop, the battery seems to last forever).
as for alpha (and yes, I do own an alpha, as well as intel and arm hardware and I'd like to have some ppc), the cost is prohibitive, especially on their new stuff... have you priced a 21264 system? ouch! especially when compared to ppc hardware.
and, with that, I'll bring a stop to my rambling...
when I was in high school I worked evenings at a local doctors' office... my friends and I had a _HUGE_ water weenie built from a length of surgical tubing I picked up an work. fully inflated it was something like 20 feet long; we coiled it in the passenger footwell in my car... that thing held _so_ much water, and lasted forever. it's a ton of fun to drive around while your friend is literally riding shotgun soaking most of the teenagers you pass!
problem being that you won't get anywhere near the necessary velocity in the projected fluid, the flame burns back up the stream faster than you can possibly shoot, hits the tank and BOOM... sort of like those idiots who try to spray lighter fluid on already _burning_ coals in a barbecue, except of course a supersoaker costs a lot more than a can of lighter fluid...
Slashdot Mirror
Mostly true. The first exception that comes to mind is X. 7.2 (even with the updates) runs XFree86 4.1.0; 7.3 includes 4.2.0.
- mark
Vice-President comes before... oh, wait!
;-)
In the end you have a company populated entirely by Veeps; imagine the efficiencies...
- mark
In trying to explain open source software to friends and family, I often run into the "you get what you pay for" line. My standard answer now goes something like this...
Open source contrasts heavily with proprietary software, where it is in fact true that you get what you pay for.
With open source software, more often than not, you get an awful lot more than you paid for. If by some fluke you actually do get what you paid for, you really have no grounds for complaint......
- mark
All fine and good, except that settlements do not establish precedent. Court _decisions_ do. In any future case, the court will look for similiar case law. If BT is really attempting to establish a precedent that they can use to go after bigger fish, the will _not_ accept a settlement offer, no matter how small or big.
- mark
describe "rumoured to suck"...
I've got an aiwa cdc-mp3, works great for me. I haven't listened to the radio during my commute in literally months. and it _really_ helps during long road trips; a cd holds a _lot_ of high-bitrate music.
the only minor annoyance is that it won't restart mid-track. for me, not such a big deal.
- mark
Cryptonomicon had a fairly good example of this. it's fictional, but the concepts apply.
as the story unfolded, the Japanese had a code that we hadn't broken. but could track when such encoded messages were sent, and what traffic they spurred. so a single message spawns many other messages, possibly using codes we have broken. in the particular instance in the story, the resulting messages went out ordering all available Japanese excavation and mining engineers to the Phillipines. hmmm... the Japanese are digging something BIG in the Phillipines. and it is important enough to them that the original orders were encrypted with this new code......
so, without decoding the original message, you can at least analyze the ripples it produces, and learn _something_ about the contents.
- mark
this is pretty much what happened to informix. they bought illustra and rolled illustra ordmbs tech into informix to create informix 9. they did an amazing job prepping their sales force for the release. so when the release slipped, they'd already convinced their users that version 9 was _so_ much better than version 7. naturally sales of the informix 7 slipped.
the story ends with a buyout by ibm. pretty much for market share, since db2 had most of the same features as informix 9.
so, customers do actually hold out for "so much better" product you've primed them for. and it certainly doesn't help your bottom line in the short term.
- mark
is that hits or pages for cnn? I count about 60 images on the cnn.com front page. still works out to an impressive ~850 _pages_ for 50k _hits_, but not as impressive as 50k pages would have been... that's the whole point of the comment earlier about cnn dropping their contract with akamai. with akamai, cnn would have been serving those 850 pages, but never would have seen the 50,000 images requests...
- mark
"first research laboratory started by a software company"
Xerox isn't a software company, at least not primarily. of course, one could argue that Microsoft is no longer exclusively a software company (keyboards, mice, xbox...), but that's another discussion. they were certainly only a software company when the founded Microsoft Research, and I believe this is the point the poster was attempting to make...
- mark
well, Ron Buckmire, one of my math professors in college did his PhD work on "The Design Of Shock-Free Transonic Slender Bodies" [abacus.oxy.edu]. pretty interesting stuff, though the math gets a bit deep.
if I remember correctly (it's been a few years since I heard him describe the work), they worked out a closed form solution to the equations that govern fluid flow over a transonic body. given that, it isn't too hard to engineer a fuselage that doesn't generate a boom as it reaches mach 1. oddly enough, it looks suspiciously like a missile.
- mark
actually, it can be a bit simpler than that... if you use qmail as the mail server for your domain, then everything up to '-' is evaluated as an address for delivery. I routinely give
from-[NameOfCompany]@[mydomain]
as my email address when I'm forced to register. all such mail is then delivered to from@[mydomain] by qmail. the advantage here is you aren't creating a million aliases to keep track of; and it is _very_ easy to see who sold my name to whom.
- mark
actually, in the mysql fulltext search docs the mysql guys do a decent job of detailing the _two_ options for querying the fulltext index. putting the "match (a,b) against ('words')" in the "select" list returns the relevance rank against all rows in the table, including those with zero relevance. if you put the match clause as part of the "where" clause then mysql returns only those rows with non-zero relevance, ordered by the relevance score (but the relevance score is not one of the return columns).
- mark
actually, since 3.23.23, mysql supports fulltext indexing of varchar and text columns in myisam tables. the results are naturally much faster than a like query against a column or columns.
indexing your archive might be a good test of their text index scalability...
I'm not sure how things have changed (it hasn't been _that_ long) but when I was in school, D's of any sort were not considered passing...
- mark
actually, the akamaized link should stay the same (e.g. a935.g.akamaitech.net/...). akamai's dynamic dns chooses the best box to serve it to you, so your squid cache should only contain one copy.
- mark
but then I started thinking about me as a developer and sites that I've built to make heavy use of services like akamai. now _none_ of my images will show up in your browser because all of the images are served out of akamaitech.net (akamai's distributed network). now it doesn't sound so great...
unless I could setup akamaitech.net as an exception, and therefore acceptable to fetch images from (but how many average joe users of netscape would actually understand why my page looked bad in the first place?). but then, if I allow akamai images, the ad banner folks will just start akamaizing their delivery (not a bad idea to begin with), and I'm back to square one...
just my (probably incoherent) ramblings...
- mark
you probably want to read this article on the real complexity involved in this sort of localization, and the pitfalls inherent in the sort of substitution you suggest.
suffice it to say, it's a nightmare for anything beyond nice and friendly iso8859-1. the author uses an amusing anecdote about a simple localization of a simple feedback message into Chinese, Arabic, Russian and Italian...
- mark
in that case email and talk are about the only way to have any meaningful and frequent interaction with family. bet they didn't take that into account, huh?
- mark (speaking from personal experience)
as I see it, the right to pursue gainful employment is always going to win out in the end.
yes, I had to sign a non-compete when I started with my current employer, but I've never worried about it. for one thing I live and work in california where (as previously mentioned) these sort of contract provisions have no teeth.
beyond that, it all comes down to how they attempt to define a "competitor". they might conceivably be able to stop you from going to work for their number one rival (and even that is a long shot here in california), but there is absolutely no way they can claim the primary business to be "web" and bar you from working for _any_ other web firm. ludicrous!
- mark
perhaps they could call it "the distribution formerly known as red hat"?
- mark
it really depends on the branch of mathematics you are interested in (of course)... I'm personally into Analysis, so I was looking at (and continue to look at)
University of British Columbia
Johns Hopkins
Columbia
though keep in mind that there are other schools that do Analysis well, these are just the top schools on my radar.
also note: last time I checked, the market for a PhD in Mathematics was abyssmal. my senior year of undergrad I was on the search committee for _one_ tenure-track position at a small (but good) school. we had to wade through ~1000 serious applicants, and I've heard that other job postings are similiarly competitive. this is why I am still in private industry...
- mark
the issue isn't just the inexpensive board, it is the cost and performance of the chip (let's _not_ start any g3 vs. piii religious wars here). and for me personally the power consumption of the chip is important as well... it might be easier to run a linux-intel box as my network gateway, but I like my netwinder because of the _way_ low power consumption. likewise the ppc chips perform much better than intel hardware in power consumption arena (ever used a g3 laptop, the battery seems to last forever).
as for alpha (and yes, I do own an alpha, as well as intel and arm hardware and I'd like to have some ppc), the cost is prohibitive, especially on their new stuff... have you priced a 21264 system? ouch! especially when compared to ppc hardware.
and, with that, I'll bring a stop to my rambling...
- mark
when I was in high school I worked evenings at a local doctors' office... my friends and I had a
_HUGE_ water weenie built from a length of surgical tubing I picked up an work. fully inflated it was something like 20 feet long; we coiled it in the passenger footwell in my car... that thing held _so_ much water, and lasted forever. it's a ton of fun to drive around while your friend is literally riding shotgun soaking most of the teenagers you pass!
- mark
I'm pretty sure it would blow up in your hands...
problem being that you won't get anywhere near the necessary velocity in the projected fluid, the flame burns back up the stream faster than you can possibly shoot, hits the tank and BOOM... sort of like those idiots who try to spray lighter fluid on already _burning_ coals in a barbecue, except of course a supersoaker costs a lot more than a can of lighter fluid...