Yeah, but it is probably getting easier now to thread electronic transfers through an untrackable maze of accounts/countries than to handle the cash.
In the UK (and accross EU), money laudering regs cover almost any large (>10k) cash transactions, requiring associated paper-trail which means you might as well have done it electronic. of course the paper trail may break down at some point (as may an electronic one) - but the great thing about these laws (for the gov) is they always get to prosecute someone, since failure to maintain/audit/check the paper trail is criminal...
So, while politicians can obviously still take a big pile of cash, it's no use if you can't spend it (safely).
Much more likely it's a convenient excessively-paid low-hours corporate advisor position at a shell company that's definitely not related to any deals done in office (I mean the Caymans are nowhere near Lichenstien so it can't be can it...).
If you are a researcher it is in your interest to [...]. Putting your articles on a website is all you need to do.
That may be the core issue that this reviewer has (but doesn't dare state). The old established journals (and by association the commercial databases of old established journals) absolutely do NOT want you putting your articles on the web.
Google Scholar's real threat is probably not that it is a worse/better/whatever academic search engine - but rather that it could add a lot of momentum to the move towards web-based scientific publishing. That is what the traditional industry (journals and databases) is really really scared of.
Actually accroding to Linus. Like I said, where do think he got his GIT data from ?
If this new GIT database is incomplete then perhaps someone should have waited until the lifeboat (ie. the export) was checked before pushing the self-destruct button.
If the GIT database is complete, then what was the problem (that the reverse engineering was solving) ?
So, is Linus' new dataset incomplete or not ? He doesn't seem to think so (anywhere I've seen) and surely he should know ?
Notice how he didn't announce he got the data with Tridge's help - why ? - because it wasn't necessary, everything could be exported anyhow. But apparently that wasn't enough for Tridge.
Re:Hold on, I need to type a message to 911...
on
Microsoft's 911 Patent
·
· Score: 2, Interesting
If the airbag deploys there is a clear emergency, and the system can press the button itself - ISTR some in-car systems already do this.
I don't doubt that Larry pushed the connection for all it was worth, just that I don't think it could have been worth a lot (particularly compared to being in use for Linux dev and improving the productivity - which came later, after BK expanded).
I mean how does the sales call go ?
"we'd like to sell you our new SCM, BK, we've got Linus Torvalds advising on it!" "hmm, I thought he famously didn't use SCM ?" "yeah that's right, he says he hasn't found one that's good enough" "so does he use this BK thing ?" "well er no, not yet" "why not ?" "well er it's not good enough - yet, but we're working on it and he says he will use it soon..." "yeah well call me when it's actually "good enough""
Removing code is not "development" and no it cant cost anywhere near as much
Heard that many times from Sales. Often they love the idea of a cheap/lite version - target people with lower budgets whilst protecting existing revenue. After all, "it can't be that hard just to remove a few menu entires can it". All good in theory, but IME it never works out like that and always costs a _lot_ more than you think when you add it up at the end.
It is my understanding that Larry worked on BitKeeper with Linus (a process which started when he was still at Sun) for a very long time just by himself. Linus and Larry are personal friends. Larry even visited at Linus' house when he was hashing out BitKeeper. Only after the project matured and became commercially viable did he get financing and other people on board.
Bitmover is verifiably several people by 1999-2000 timeframe, and it doesn't look to me like Linus actually started using it (as opposed to saying he might when it worked) until late 1999 at the earliest.
For the product to have a measurable effect and acquire a reputation based on Linus' use of it would take several months at least, so that would be well into 2000 - which is after bitmover grew. So, either it grew commercially viable before Linus' used it (and hence not off his reputation), or the dev investment up front was several people not one.
Note that this doesn't contradict that Linus helped Larry on BK well before that time - but it is the reputation of BK being "used by Linus", not that "Linus helped the people writing it" that is being claimed as valuable payback.
This airshow was staged badly (Ukraine I think - they still flew over/towards the crowd after western shows stopped doing that).
1-2 tons, where did you get that from ? This thing is an ultralight - 254lb plus 5gal fuel plus 350lb max load, so only about 600lb in all.
The SU 27 is 20-30 tons, with 5+ tons of fuel (vs. 5 gallons), probably travelling double the speed this airscooter thing is capable of. Oh, and it did crash into an "other gathering" - it crashed into the crowd watching the show.
Try doing your math with a 40 tonne truck travelling at 60mph. How many people would one of those kill if it left the road and ploughed through a school ? Lots more than 75 by your logic. I say not.
You are also way out on a hand grenade in a crowd - people have done that before and the body counts are way below 75.
If you want to kill 75+ people with either a hand grenade or an airscooter, you need to stick them in the path of an airliner. _That_ is where your problem is (with all ultralights) - not with them flying into the ground.
I think it would be really hard to kill 75 people crashing one of these, particularly if it was out of gas (nothing to burn then).
I think the worst air-show crash was 83 dead which was an SU-27 into a crowd - that's a lot bigger aircraft, with a lot more fuel, going a lot faster into a dense crowd.
Not arguing about why he did or did not spend money, simply how much and on what.
I can see larry's figure of 500k to develop support and host the free version (and I do mean just the extras / differences for the free version) being reasonable. IME just hosting (properly on decent dedicated servers) a complex web app for a few hundred users can run to 1000s per month.
In contrast, suggesting 500k is the cost of the whole of bitkeeper dev (or even "the vast bulk" of it) seems like the ridiculous number to me. I can't see it being anywhere near that cheap if it was done in the US (or UK/EU etc.) - maybe if you offshored all of it somewhere really cheap it would be possible.
I'd be interested in what you think is the bitmover timeline - eg. when Larry employed others and when Linus started using it. I don't have personal knowledge but web/usenet history seems to show that it was larry+others _before_ Linus switched to it.
Some people have short memories, from just a little up the thread: Now look at this ridiculous numer of $500k... what an ass! Thats the money he spent on the whole R&D effort of his product which he goes and sells for profit
and now: Remember, noone is arguing that the entire development cost of BitKeeper was under $500k. It is in fact quite likely that it was more
No sense continuing to debate when I can just watch you argue with yourself.
Yep, the whole linux user community (which last I looked wasn't just Larry) has received far more value from Linus than $500k.
Each Linux user has not received that total value though - I haven't (could have bought commercial Unix for orders of magnitude less in total).
Similarly, if BK (as they claim) improved just Linus' (let alone anyone else's) productivity by a factor of 2 for a period of a few years, then the benefit from BK just on Linus time-costs is also close to the 500k figure if not more. But again, that is the benefit to the whole community, not one person.
As has been said before. If the server allows the client to corrupt data then it is the server that is broken, not the client.
Yep, I've heard that too, and I think it could be wrong.
The user tells the client what to do, the client tells the server what to do, and a broken client could tell the server to do something the user did not intend. Eg. the user checks in a file which was shared, the client tells the server to branch and checkin on one branch, or maybe it just tells it to delete instead. Oops. Now run that across thousands of files in multiple branches. Sure, you didn't "corrupt" the SCM database - it is still there and consistent -, but boy did you mess up the data.
It is like saying that, oh I dunno, say: a webdav client shouldn't be allowed to "corrupt" a webdav server. Whilst also saying that a webdav server must implement delete operations. A broken client can do a delete which wasn't intended. Is that the server's fault ?
I do have experience since my consulting firm at one point was involved in custom coding. We had 6 programmers at the peak of that project. How about you?
If it matters, 10+yrs in dev with 5+ yrs leading small (5-10) dev teams, for last few years CTO of company of 20-30 people (at it's peak).
Bitmover currently has 8 vacancy ads, some of which are for multiple positions. Four "founders" are listed.
Even if we assume that is all smoke & mirrors and it is just Larry, with no office or any other overheads, that $500k from 1997 works out at less than $100k per year. Then you have to take out healthcare & other employer expenses for a US company - not insignificant. Quick google for a relevant salary survey shows you end up with less than one programmer for that. So it _still_ can't be only 500k even if it was just Larry all the time.
"Just an SCM system" doesn't mean a lot - like saying "just a database". In some people's opinion it seems to be a pretty good scm and better than a lot of others. Wonder how big the sourcesafe dev team is ? I'd bet on more than $500k _per year_, and sourcesafe sucks.
$500k is all of the costs for his product ? Obviously false, obviously ridiculous if you had _any_ experience of paid-for software development. Words fail me.
Even if that was _just_ programmer costs (and not the whole company costs - given that this is a one product company) it would add up to only 2-3 man years. So that's less than one programmer through the company/product history.
Bitmover has _several_ programmer _vacancies_ advertised right now, and yet you reckon they do it all with one part time.
I'm not "falling for" anything - you just can't do simple math.
500k is 2-3 man years at my guess (I'm not in the US and don't know US salaries but I can look up salary surveys and add estimated overheads).
Even if it was "merely a stripped-down, crippled version", just doing the testing and release engineering costs time and money. Then add in the support costs and dev and support costs for all the extras added in (eg. data export & integration to other scms) for the open source community, and the server hosting & support... 2-3 man years doesn't seem high to me.
There is also no way that 500k is "all the work on ButtKeeper". Based on the info on their site, I would guess that bitmover is 20+ people (could well be more), and they've been around for a few years. At that size my guess is that you are easily into costs of $millions _per year_ - and bitkeeper is all they do. Do the math.
Look at the title of your post, read LInus' writings on this, and it is clear that you (among many others) are completely wrong.
Data lock-in is not an issue here. Full meta-data export was already written (where do you think Linus got his data for his new system, that he is now writing instead of writing the kernel, from ?). Export for any format could have and would haave been put in by Larry, and according to Linus this was "not enough" for Tridgell.
So, if data lock-in (a valid concern and obviously a good reason to talk to your vendor and agree export routes and formats...) was not the issue, what was it ?
Slashdot Mirror
Yeah, but it is probably getting easier now to thread electronic transfers through an untrackable maze of accounts/countries than to handle the cash.
In the UK (and accross EU), money laudering regs cover almost any large (>10k) cash transactions, requiring associated paper-trail which means you might as well have done it electronic. of course the paper trail may break down at some point (as may an electronic one) - but the great thing about these laws (for the gov) is they always get to prosecute someone, since failure to maintain/audit/check the paper trail is criminal...
So, while politicians can obviously still take a big pile of cash, it's no use if you can't spend it (safely).
Much more likely it's a convenient excessively-paid low-hours corporate advisor position at a shell company that's definitely not related to any deals done in office (I mean the Caymans are nowhere near Lichenstien so it can't be can it...).
except concord cruised at that sort of height too - and the sonic booms definitely were a problem
Eh ? Why the F**k can't they use CGI like every other damn movie these days ?
If you are a researcher it is in your interest to [...]. Putting your articles on a website is all you need to do.
That may be the core issue that this reviewer has (but doesn't dare state). The old established journals (and by association the commercial databases of old established journals) absolutely do NOT want you putting your articles on the web.
Google Scholar's real threat is probably not that it is a worse/better/whatever academic search engine - but rather that it could add a lot of momentum to the move towards web-based scientific publishing. That is what the traditional industry (journals and databases) is really really scared of.
"So why doesn't she sue her ex-boyfriend"
- at a guess, because he doesn't _have_ $3 million. If he did, he probably wouldn't be ex.
Actually accroding to Linus. Like I said, where do think he got his GIT data from ?
If this new GIT database is incomplete then perhaps someone should have waited until the lifeboat (ie. the export) was checked before pushing the self-destruct button.
If the GIT database is complete, then what was the problem (that the reverse engineering was solving) ?
So, is Linus' new dataset incomplete or not ? He doesn't seem to think so (anywhere I've seen) and surely he should know ?
The version history was never "lost".
What do you think Linus is using to test GIT ?
Notice how he didn't announce he got the data with Tridge's help - why ? - because it wasn't necessary, everything could be exported anyhow. But apparently that wasn't enough for Tridge.
If the airbag deploys there is a clear emergency, and the system can press the button itself - ISTR some in-car systems already do this.
I don't doubt that Larry pushed the connection for all it was worth, just that I don't think it could have been worth a lot (particularly compared to being in use for Linux dev and improving the productivity - which came later, after BK expanded).
I mean how does the sales call go ?
"we'd like to sell you our new SCM, BK, we've got Linus Torvalds advising on it!"
"hmm, I thought he famously didn't use SCM ?"
"yeah that's right, he says he hasn't found one that's good enough"
"so does he use this BK thing ?"
"well er no, not yet"
"why not ?"
"well er it's not good enough - yet, but we're working on it and he says he will use it soon..."
"yeah well call me when it's actually "good enough""
Removing code is not "development" and no it cant cost anywhere near as much
Heard that many times from Sales. Often they love the idea of a cheap/lite version - target people with lower budgets whilst protecting existing revenue. After all, "it can't be that hard just to remove a few menu entires can it". All good in theory, but IME it never works out like that and always costs a _lot_ more than you think when you add it up at the end.
It is my understanding that Larry worked on BitKeeper with Linus (a process which started when he was still at Sun) for a very long time just by himself. Linus and Larry are personal friends. Larry even visited at Linus' house when he was hashing out BitKeeper. Only after the project matured and became commercially viable did he get financing and other people on board.
Bitmover is verifiably several people by 1999-2000 timeframe, and it doesn't look to me like Linus actually started using it (as opposed to saying he might when it worked) until late 1999 at the earliest.
For the product to have a measurable effect and acquire a reputation based on Linus' use of it would take several months at least, so
that would be well into 2000 - which is after bitmover grew. So, either it grew commercially viable before Linus' used it (and hence not off his reputation), or the dev investment up front was several people not one.
Note that this doesn't contradict that Linus helped Larry on BK well before that time - but it is the reputation of BK being "used by Linus", not that "Linus helped the people writing it" that is being claimed as valuable payback.
It must not weigh at least 1500lbs - in fact as an ultralight it _has_ to weigh less than 254lbs unladen (be around 600 max loaded as I read it).
The Robinson is nearer 1500 - much more than twice the weight.
This airshow was staged badly (Ukraine I think - they still flew over/towards the crowd after western shows stopped doing that).
1-2 tons, where did you get that from ? This thing is an ultralight - 254lb plus 5gal fuel plus 350lb max load, so only about 600lb in all.
The SU 27 is 20-30 tons, with 5+ tons of fuel (vs. 5 gallons), probably travelling double the speed this airscooter thing is capable of.
Oh, and it did crash into an "other gathering" - it crashed into the crowd watching the show.
Try doing your math with a 40 tonne truck travelling at 60mph. How many people would one of those kill if it left the road and ploughed through a school ? Lots more than 75 by your logic. I say not.
You are also way out on a hand grenade in a crowd - people have done that before and the body counts are way below 75.
If you want to kill 75+ people with either a hand grenade or an airscooter, you need to stick them in the path of an airliner. _That_ is where your problem is (with all ultralights) - not with them flying into the ground.
The airscooter (which is what the article is about) is an ultralight helicopter with co-axial rotors.
See eg.
image link
Where does it say it is a rotary engine ?
It _does_ say lots about four-stroke and pistons and cylinders - all of which sounds nothing like a rotary engine to me.
I think it would be really hard to kill 75 people crashing one of these, particularly if it was out of gas (nothing to burn then).
I think the worst air-show crash was 83 dead which was an SU-27 into a crowd - that's a lot bigger aircraft, with a lot more fuel, going a lot faster into a dense crowd.
From the article, the engine is only 65hp, so fuel consumption may be better than some cars.
It also says 2hrs flight time, 55knots (approx 60mph), on 5 gallons. That is >20mpg, which would definitely be better than the worst SUVs.
Not arguing about why he did or did not spend money, simply how much and on what.
I can see larry's figure of 500k to develop support and host the free version (and I do mean just the extras / differences for the free version) being reasonable. IME just hosting (properly on decent dedicated servers) a complex web app for a few hundred users can run to 1000s per month.
In contrast, suggesting 500k is the cost of the whole of bitkeeper dev (or even "the vast bulk" of it) seems like the ridiculous number to me. I can't see it being anywhere near that cheap if it was done in the US (or UK/EU etc.) - maybe if you offshored all of it somewhere really cheap it would be possible.
I'd be interested in what you think is the bitmover timeline - eg. when Larry employed others and when Linus started using it. I don't have personal knowledge but web/usenet history seems to show that it was larry+others _before_ Linus switched to it.
Some people have short memories, from just a little up the thread:
Now look at this ridiculous numer of $500k... what an ass! Thats the money he spent on the whole R&D effort of his product which he goes and sells for profit
and now:
Remember, noone is arguing that the entire development cost of BitKeeper was under $500k. It is in fact quite likely that it was more
No sense continuing to debate when I can just watch you argue with yourself.
Yep, the whole linux user community (which last I looked wasn't just Larry) has received far more value from Linus than $500k.
Each Linux user has not received that total value though - I haven't (could have bought commercial Unix for orders of magnitude less in total).
Similarly, if BK (as they claim) improved just Linus' (let alone anyone else's) productivity by a factor of 2 for a period of a few years, then the benefit from BK just on Linus time-costs is also close to the 500k figure if not more. But again, that is the benefit to the whole community, not one person.
As has been said before. If the server allows the client to corrupt data then it is the server that is broken, not the client.
Yep, I've heard that too, and I think it could be wrong.
The user tells the client what to do, the client tells the server what to do, and a broken client could tell the server to do something the user did not intend. Eg. the user checks in a file which was shared, the client tells the server to branch and checkin on one branch, or maybe it just tells it to delete instead. Oops. Now run that across thousands of files in multiple branches. Sure, you didn't "corrupt" the SCM database - it is still there and consistent -, but boy did you mess up the data.
It is like saying that, oh I dunno, say: a webdav client shouldn't be allowed to "corrupt" a webdav server. Whilst also saying that a webdav server must implement delete operations. A broken client can do a delete which wasn't intended. Is that the server's fault ?
One (decent, experienced) programmer since 1997 adds up to more than $500k by my reckoning, just on salary before adding _any_ overheads.
Doesn't matter if I take UK salaries and convert to $ or look up US salary surveys - it's still more.
I do have experience since my consulting firm at one point was involved in custom coding. We had 6 programmers at the peak of that project. How about you?
If it matters, 10+yrs in dev with 5+ yrs leading small (5-10) dev teams, for last few years CTO of company of 20-30 people (at it's peak).
Bitmover currently has 8 vacancy ads, some of which are for multiple positions. Four "founders" are listed.
Even if we assume that is all smoke & mirrors and it is just Larry, with no office or any other overheads, that $500k from 1997 works out at less than $100k per year. Then you have to take out healthcare & other employer expenses for a US company - not insignificant. Quick google for a relevant salary survey shows you end up with less than one programmer for that. So it _still_ can't be only 500k even if it was just Larry all the time.
"Just an SCM system" doesn't mean a lot - like saying "just a database". In some people's opinion it seems to be a pretty good scm and better than a lot of others. Wonder how big the sourcesafe dev team is ? I'd bet on more than $500k _per year_, and sourcesafe sucks.
$500k is all of the costs for his product ? Obviously false, obviously ridiculous if you had _any_ experience of paid-for software development. Words fail me.
Even if that was _just_ programmer costs (and not the whole company costs - given that this is a one product company) it would add up to only 2-3 man years. So that's less than one programmer through the company/product history.
Bitmover has _several_ programmer _vacancies_ advertised right now, and yet you reckon they do it all with one part time.
I'm not "falling for" anything - you just can't do simple math.
500k is 2-3 man years at my guess (I'm not in the US and don't know US salaries but I can look up salary surveys and add estimated overheads).
Even if it was "merely a stripped-down, crippled version", just doing the testing and release engineering costs time and money. Then add in the support costs and dev and support costs for all the extras added in (eg. data export & integration to other scms) for the open source community, and the server hosting & support... 2-3 man years doesn't seem high to me.
There is also no way that 500k is "all the work on ButtKeeper". Based on the info on their site, I would guess that bitmover is 20+ people (could well be more), and they've been around for a few years. At that size my guess is that you are easily into costs of $millions _per year_ - and bitkeeper is all they do. Do the math.
Look at the title of your post, read LInus' writings on this, and it is clear that you (among many others) are completely wrong.
Data lock-in is not an issue here. Full meta-data export was already written (where do you think Linus got his data for his new system, that he is now writing instead of writing the kernel, from ?). Export for any format could have and would haave been put in by Larry, and according to Linus this was "not enough" for Tridgell.
So, if data lock-in (a valid concern and obviously a good reason to talk to your vendor and agree export routes and formats...) was not the issue, what was it ?