"At the Hot Chips conference at Stanford University, Sun presented plans for a security accelerator chip that it said would reduce encryption costs for applications such as VoIP calls and online banking Web sites. The coprocessor will be included on the same silicon as Rainbow Falls, the code name for the follow-on to Sun's multi-threaded Ultrasparc T2 processor."
Any experienced buzzword bingo player should have shouted out before reaching the end of the first sentence.
It is bloody fucking lot more than anybody else can claim.
While this is still an achievement and i can honor the amount of work involved (i worked myself silly by proving 7 lines of code during my university times), my comment refers to the phrase "World's First Formally-Proven OS Kernel". Most people think of "proven" as "proven to be correct" (which isn't the same as i tried to point out). Indeed in the university, we had a project at which was named "Procos" (Proven correct systems) to which i always referred to "Probably corrects systems". While i believe that a "proven" system will be of a higher quality than an unproven system, the costs involved usually are to high for the amount of quality increase. My impression is, that the same quality increase can be obtained at much lower cost.
But my philosophical concerns go even deeper. As in this case, software is used to prove the code. In which case, the process of proving is a glorified version of compiling. You start out with a formalized spec and a piece of code. By proving you eliminate the differences between them. Which means you're using software to make sure the output matches the specs. Which is what a compiler does. I'm aware that this is oversimplifying. But proving is (in my eyes) compiling with a manual two way crosscheck of dubious parts.
CU, Martin
P.S. I'm licensing all spelling errors under the conditions of the LGPL.
To prove that a specification matches an implementation 100%, the specification needs to be highly formalized. In this case it was Haskell, which is a functional programming language. It may be not as similar as Perl and Python are, but my statement still stands.
There is an old corollary that says, you cannot get from the informal to the formal by formal means. All they have proven is, that two specifications contain the same bugs. Both specification were formal (Haskell, C). This is the same as having Perl and Python code and you to prove they implement the same functionality. Neither is a proof, it is bug free (informal definition of bug, not if a bug is specified it isn't a bug).
This couldn't happen in the western world... Those camps would be at least named bootcamp. You wouldn't be beaten to death, they just make you run until you drop dead.
i don't think that this an atypical problem, neither inside or outside the
open source community. We have people giving ressources of to projects
(e.g. time, money). Usually they expect something in return (e.g. recognition,
influence). Normally those expectations are never stated explecitely. So what
happens: Someone sees his expectations not met, so he cuts the ressources he
gives. Usually this goes together with hurt feelings as well, so he tries to
get a refund by keeping assets (domains, money, passwords, etc.).
Same thing happened with other OSS projects (e.g. Blastwave) and non
profit organisations (e.g. Hannelore Kohl Stiftung here in germany).
You cannot fix this. When you try to fix it, you need a board and a charta
right at the beginning. Too many projects would already die here and would
never get to the stage where a quitting founder brings a crisis. In the worst
case now: they have to start at the current status again under a new name.
CU, Martin
P.S. This shell not be a factual description, what happened in this project.
This is only a description of things i observed elsewhere and would expect
to find here too.
my best guess is, that the attack angle will shift soon. Instead of decrypting the content of the disk, there will be software/hardware to break the HDCP-protection. It would be a lot more difficult close holes there, as thousands of devices are already out (which cannot be modified but only blacklisted).
when Princess Diana died in 1997, we were supplying support services for one of the biggest news sites here in germany. It hit the site like a Tsunami. Unluckily someone reported in an IRC channel, that the news site would display pictures of the dying princess. So there was a real frenzy. It started early in the morning and we were called to fix a server malfunction. Unluckily the server malfunction turned out to be 99+% TCP SYN packets on the incoming side of the internet connect. That was at a time, when major news sites were connected by 2mbps lines:-). We were so fixed on locating a technical problem, it took us some minutes to connect the symptons to an event in the real world. Luckily the cab driver who picked me up had his radio on.
i think Seneca leftt us with good quote for this: "Timeo Danaos et dona ferentes" (I fear the Danaans (Greeks) even if they bring gifts). This might become a mighty bumerang if he gets convicted.
I can clearly see the advantages for Tauss: If he gets convicted, he hasn't lost anything. If not, he may emerge a leader in a new party. The advantages for the Pirate party are clearly limited: Tauss seat expires in September and if he gets convicted, they'll take a mighty pounding.
But the pirate party has little choice in that matter. As long as he hans't been convicted, he is free to join any party of his choice. And his publicly stated opinions on Internet topics were very close to the pirate party lines.
my personal opinion is, that BIND 9 already lived too long and BIND 10 started much too late. If you have to operate huge installations (>250.000 Zones), BIND 9 is close to unuseable.
Example: Starting BIND 9 with 350.000 Zones already consumes the complete service window (2 hours) we have for works concerning the hardware. You can't even shave off much time by having all zone files on a ram disk (about 10% less time). BIND 9.6 utilizes a single core for 2 hours just to parse and load the information. For comparison a different (comercial) product imports the (same) complete configuration in about 90s (from disk, BIND 9 format) and takes about 4s for start afterwards. I know there are workarounds for BIND, but they come with high operational costs.
BIND is (IMHO) mainly a reference implementation. It has to implement everything in one single product and suffers the usual penalties for it. I still use BIND 9 myself for several purposes since it has a some advantages too (mainly, that it is OSS).
Sincerely yours, Martin
P.S. If there is any interest, i can post some benchmarks and scripts which i used to run them....
DISCLAIMER: I'm working for a company that is selling DNS products. So i'm not to be considered a neutral party:-). But since i'm doing this for 15 years now, i consider myself at least an experienced biased party.....
Most people will have emotional problems of trusting a computer that may have a glitch. But you have to be aware that you're trusting your live regularely to other people who are drunk, bleary-eyed or working beyond their capability (not always, but take sometimes a look at your bus/cab driver). For me the decision is not to be based on emotions but on statistics.
my personal opinion is, that it comes down to statistics. Do human errors cause more accidents (when flying planes) or computer glitches? Does the manual override prevent crashes or does it help the pilots to cause more? Without claiming to be an authority on the topic, my impression is, that the statistics are clearly on the side of the computers.
I would rather take a 0,0001% probabilty to be killed by a computer glitch than 0,0002% probability to be killed by human error (figures are no real numbers, they just serve as an example). Be aware that an override doesn't just work for the genius pilot with catlike reflexes and 10.000 hours of flight experience. It also works for the rookie, inattentive or ill trained one (example for such an accident probably caused by its crew is Colgan Air 3407).
Most people will have emotional problems of trusting a computer that may have a glitch. But you have to be aware that you're trusting your live regularely to other people who are drunk, bleary-eyed or working beyond their capability (not alwa
the pirate party reached in germany 0,9%. Concerning lack of attention from the media, nearly non-existent funds and that stupid name, this is a very strong result for them.
this explanation is clearly a case for Occam's Rasor. There are currently more simple and more probable explanations than a meteor strike. So unless other evidence comes up, this theory should rest in peace until then.
I had to solve such a problem once for a customer of us. A domain expired by accident and fell into the hands of a domainsquatter. The poor ex-owner had already advertisement material printed with his domain name on it. Damages would have ranged at about 10K$.
The problem: If a german company tries to purchase the domain, the prices tend to skyrocket (probably the same for US companies). So we created a fake russian student (not very rich) who wanted to use the domain for his private web site. He had a russian email address, had a small home page with his russian ISP etc. This way with a little negotiation, we managed to purchase the domain at a very reasonable price.
You have to be careful to become the owner of the domain. At first they tried to "lease" the domain to us by just setting the records. But it was completely in accordance with our virtual pesonality to display some paranoia and insist on a complete domain transfer.
I like to use PERT estimates, or something similar:
Essentially you are correct. But in this case the question was from someone new to project management. From the conext of the question i took, that those managed by him are as new to being project managed by him as he is project managing them. In that case we probably two effects:
They have a lot of difficulties guessing the time needed for a task in an environment where they have to cooperate with others. As a result the time waiting for input and the dependencies from others will be massively underestimated.
Most tasks will be small tasks where the answers will be something like 1 hour to 2 days. Estimates for such tasks will usually be off by an incredible factor.
As a result my formula (if you want to call it like that) will work at first better than expected. If you learn to take estimates from others and they learn to make them, you may move to your formula as it will work better on the long term
Even as my formula started out as joke once, unluckily it has proven its "value" more than once.
Slashdot Mirror
"At the Hot Chips conference at Stanford University, Sun presented plans for a security accelerator chip that it said would reduce encryption costs for applications such as VoIP calls and online banking Web sites. The coprocessor will be included on the same silicon as Rainbow Falls, the code name for the follow-on to Sun's multi-threaded Ultrasparc T2 processor."
Any experienced buzzword bingo player should have shouted out before reaching the end of the first sentence.
Current rating:
40% Redundant
30% Insightful
30% Overrated
Can't make your mind up, eh? :-)
CU, Martin
All they have proven
It is bloody fucking lot more than anybody else can claim.
While this is still an achievement and i can honor the amount of work involved (i worked myself silly by proving 7 lines of code during my university times), my comment refers to the phrase "World's First Formally-Proven OS Kernel". Most people think of "proven" as "proven to be correct" (which isn't the same as i tried to point out). Indeed in the university, we had a project at which was named "Procos" (Proven correct systems) to which i always referred to "Probably corrects systems". While i believe that a "proven" system will be of a higher quality than an unproven system, the costs involved usually are to high for the amount of quality increase. My impression is, that the same quality increase can be obtained at much lower cost.
But my philosophical concerns go even deeper. As in this case, software is used to prove the code. In which case, the process of proving is a glorified version of compiling. You start out with a formalized spec and a piece of code. By proving you eliminate the differences between them. Which means you're using software to make sure the output matches the specs. Which is what a compiler does. I'm aware that this is oversimplifying. But proving is (in my eyes) compiling with a manual two way crosscheck of dubious parts.
CU, Martin
P.S. I'm licensing all spelling errors under the conditions of the LGPL.
A specification is not an implementation
To prove that a specification matches an implementation 100%, the specification needs to be highly formalized. In this case it was Haskell, which is a functional programming language. It may be not as similar as Perl and Python are, but my statement still stands.
Sincerely yours, Martin
There is an old corollary that says, you cannot get from the informal to the formal by formal means. All they have proven is, that two specifications contain the same bugs. Both specification were formal (Haskell, C). This is the same as having Perl and Python code and you to prove they implement the same functionality. Neither is a proof, it is bug free (informal definition of bug, not if a bug is specified it isn't a bug).
The need is there, no doubt. But need does not equate success...
This means, you can be forced to do self-incrimination. What's next? Do we remove the right to remain silent? In dubio contra reo?
This couldn't happen in the western world... Those camps would be at least named bootcamp. You wouldn't be beaten to death, they just make you run until you drop dead.
Nobody really expected an expansion for Sims 3... This comes as a complete suprise. Gamers are in an uproar...
Let's face it, System Administrators get no respect 364 days a year
Let's face it and make it 365 or 365.249 days in an average year.
CU, Martin
Hi,
i don't think that this an atypical problem, neither inside or outside the
open source community. We have people giving ressources of to projects
(e.g. time, money). Usually they expect something in return (e.g. recognition,
influence). Normally those expectations are never stated explecitely. So what
happens: Someone sees his expectations not met, so he cuts the ressources he
gives. Usually this goes together with hurt feelings as well, so he tries to
get a refund by keeping assets (domains, money, passwords, etc.).
Same thing happened with other OSS projects (e.g. Blastwave) and non
profit organisations (e.g. Hannelore Kohl Stiftung here in germany).
You cannot fix this. When you try to fix it, you need a board and a charta
right at the beginning. Too many projects would already die here and would
never get to the stage where a quitting founder brings a crisis. In the worst
case now: they have to start at the current status again under a new name.
CU, Martin
P.S. This shell not be a factual description, what happened in this project.
This is only a description of things i observed elsewhere and would expect
to find here too.
Hi,
:-).
please tell me: How many time has transparent aluminium been discovered by now?
I think about five to six times... E.g. in 2005
Please don't wake me up the next time someone discovers it
CU, Martin
Hi,
my best guess is, that the attack angle will shift soon. Instead of decrypting the
content of the disk, there will be software/hardware to break the HDCP-protection.
It would be a lot more difficult close holes there, as thousands of devices are
already out (which cannot be modified but only blacklisted).
CU, Martin
Hi,
when Princess Diana died in 1997, we were supplying support services for one of the biggest news sites here in germany. It hit the site like a Tsunami. Unluckily someone reported in an IRC channel, that the news site would display pictures of the dying princess. So there was a real frenzy. It started early in the morning and we were called to fix a server malfunction. Unluckily the server malfunction turned out to be 99+% TCP SYN packets on the incoming side of the internet connect. That was at a time, when major news sites were connected by 2mbps lines :-). We were so fixed on locating a technical problem, it took us some minutes to connect the symptons to an event in the real world. Luckily the cab driver who picked me up had his radio on.
CU, Martin
Hi,
i think Seneca leftt us with good quote for this: "Timeo Danaos et dona ferentes"
(I fear the Danaans (Greeks) even if they bring gifts). This might become a mighty
bumerang if he gets convicted.
I can clearly see the advantages for Tauss: If he gets convicted, he hasn't lost
anything. If not, he may emerge a leader in a new party. The advantages for the
Pirate party are clearly limited: Tauss seat expires in September and if he gets
convicted, they'll take a mighty pounding.
But the pirate party has little choice in that matter. As long as he hans't been
convicted, he is free to join any party of his choice. And his publicly stated
opinions on Internet topics were very close to the pirate party lines.
Sincerely yours, Martin
Hi,
my personal opinion is, that BIND 9 already lived too long and BIND 10 started much too late. If you have to operate huge installations (>250.000 Zones), BIND 9 is close to unuseable.
Example: Starting BIND 9 with 350.000 Zones already consumes the complete service window (2 hours) we have for works concerning the hardware. You can't even shave off much time by having all zone files on a ram disk (about 10% less time). BIND 9.6 utilizes a single core for 2 hours just to parse and load the information. For comparison a different (comercial) product imports the (same) complete configuration in about 90s (from disk, BIND 9 format) and takes about 4s for start afterwards. I know there are workarounds for BIND, but they come with high operational costs.
BIND is (IMHO) mainly a reference implementation. It has to implement everything in one single product and suffers the usual penalties for it. I still use BIND 9 myself for several purposes since it has a some advantages too (mainly, that it is OSS).
Sincerely yours, Martin
P.S. If there is any interest, i can post some benchmarks and scripts which i used to run them....
DISCLAIMER: I'm working for a company that is selling DNS products. So i'm not to be considered a neutral party :-). But since i'm doing this for 15 years now, i consider myself at least an experienced biased party.....
Hi,
one sentence was incomplete:
Most people will have emotional problems of trusting a computer that may have a glitch. But you have to be aware that you're trusting your live regularely to other people who are drunk, bleary-eyed or working beyond their capability (not always, but take sometimes a look at your bus/cab driver). For me the decision is not to be based on emotions but on statistics.
Yours, Martin
Hi,
my personal opinion is, that it comes down to statistics. Do human errors cause more accidents (when flying planes) or computer glitches? Does the manual override prevent crashes or does it help the pilots to cause more? Without claiming to be an authority on the topic, my impression is, that the statistics are clearly on the side of the computers.
I would rather take a 0,0001% probabilty to be killed by a computer glitch than 0,0002% probability to be killed by human error (figures are no real numbers, they just serve as an example). Be aware that an override doesn't just work for the genius pilot with catlike reflexes and 10.000 hours of flight experience. It also works for the rookie, inattentive or ill trained one (example for such an accident probably caused by its crew is Colgan Air 3407).
Most people will have emotional problems of trusting a computer that may have a glitch. But you have to be aware that you're trusting your live regularely to other people who are drunk, bleary-eyed or working beyond their capability (not alwa
Sincerely yours, Martin
Hi,
the next german Bundestag will be elected in Autumn. Since a party has to pass
5% to get any seats, the Pirate Party has still a long way to go
CU, Martin
Hi,
the pirate party reached in germany 0,9%. Concerning lack of attention from the media, nearly non-existent funds and that stupid name, this is a very strong result for them.
CU, Martin
Hi,
this explanation is clearly a case for Occam's Rasor. There are currently more simple and more probable explanations than a meteor strike. So unless other evidence comes up, this theory should rest in peace until then.
CU, Martin
Hi,
Has probably something to do with detecting watermarks.... At least they're fond to set new "watermarks" everywhere.
Yours, Martin
Working by rules would have taken too long. And if you count the time spent for forms etc., it would probably cost more money too.
Yours, Martin
Hi,
I had to solve such a problem once for a customer of us. A domain expired by accident and fell into the hands of a domainsquatter. The poor ex-owner had already advertisement material printed with his domain name on it. Damages would have ranged at about 10K$.
The problem: If a german company tries to purchase the domain, the prices tend to skyrocket (probably the same for US companies). So we created a fake russian student (not very rich) who wanted to use the domain for his private web site. He had a russian email address, had a small home page with his russian ISP etc. This way with a little negotiation, we managed to purchase the domain at a very reasonable price.
You have to be careful to become the owner of the domain. At first they tried to "lease" the domain to us by just setting the records. But it was completely in accordance with our virtual pesonality to display some paranoia and insist on a complete domain transfer.
Sincerely yours, Martin
Essentially you are correct. But in this case the question was from someone new to project management. From the conext of the question i took, that those managed by him are as new to being project managed by him as he is project managing them. In that case we probably two effects:
As a result my formula (if you want to call it like that) will work at first better than expected. If you learn to take estimates from others and they learn to make them, you may move to your formula as it will work better on the long term
Even as my formula started out as joke once, unluckily it has proven its "value" more than once.
CU, Martin