Couple of points, some exploits don't work against a debug build system, that do in release mode - ever tried to track down a bug that was unreproducable when built in debug?
For another off the wall point, what if this is not actually running w2k, but Linux, and MS use the Anti-Ms brigade, to poke hole in Linux's default security?
One of the rules of engagement is that you have to tell them how you did it, but it doesn't give a time limit on when you have to tell them.
--
Would MSN messenger now be illegal?
on
UCITA is passed
·
· Score: 1
If reverse engineering is not allowed, would that now cause a problem for MicroSoft, since they reverse engineered the AIM client?
Seriously, without 100% open standards, outlawing reverse engineering is simple anticompetitive. --
If you read the article, it says you get a month's equilavent of your original machines power.
so, a month of an average AMD350 is 30*24*60 = 43,200 AMDMinutes which is worth 43,200/10757 = 4.01 AmdMinutes.
So, the proud donater of a 386SX25, could get of the magnitude of 1 second:)
--
What could be the worst solution...
on
Austria Bans Spam
·
· Score: 1
>I've even seen resistance to Corporate LAN e-mail systems such as Exchange or Notes precisely because senders are authenticated. >(For example, if a secretary sends a message from the Boss, the message will read From:Boss Sent By:Secretary = Boss gets mad because secretary can't impersonate him/her)
(Actually, in exchange they can do this, depending on the setting. You can either give the secretary "Send on Behalf of" privs which will do as you say or "Send As" privs, which will allow the secretary to completely act like the boss)
Speaking of Exchange, didn't MS say they were targeting ISP's with the next version of exchange (Platinum?) Imagine the possible (worst case) timeline...
Phase (1) MS get a few major ISP's to use exchange as email backbone, supporting "legecy" SMTP support, but advanatges for intra exchange-enabled sites (user verification with NT Challege/response etc)
Phase (2) MS add more features which only work intra-exchange sites. ISP-Exchange clients now installed as default on all windows OS's
Phase (3) Problems sending/recieving to SMTP sites, and due to critical mass being achieved by ISP-exchange, (as all the windows pre-installed clients don't now work with SMTP) it is the SMTP sites that have to adapt.
Unlikely, yes, but MS would be one of the few companys that could have both the will, and the market clout to decommotise email protocols
SMTP may be old, and have weaknesses, but it is open. If we are not careful and plan an open upgrade path (past ESTMP, and to something better and different) someone else will. --
please meet virtual machine A and B, who have been configured to do a reverse open relay check on RCPT, One a particular day, A wants to send some email to B...
A: "HELO, B" B: "pleased to meet you. A" A: "I have some mail for you" B: (thinking, do I trust A? Let me see...)
(B starts working in the background) B: HELO A A: Pleased to meet you B B: I have some email for you (hee hee, but I'm not going to send it, just checking your answer" A: (thinking, do I trust B? let me see...)
(A starts working in ther background) A: HELO B B: Pleased to meet you again, A. Busy today, isn't it? ....
In short, it's a nice idea, but it would set up an endless loop.
Thank you for publishing some of the correspondence you have recieved from some vocal linux users. What I find interesting that all of the email you have displayed have been offensive and immature. While any reasonable person will admit that in any mass of people there are extremes in any direction, the same reasonable people would not infer the behaviour of the group form the actions of a minority.
Your inital tests raised several issues (and found some bugs, now fixed - thanks!) about the imbalance between the setup for NT and Linux, details of which can be found on any reputable news source, so I shouldn't need to reiterate them here. The general conclusion you attempted to dervive from that specific setup (NT is x times faster than Linux) was at best flawed, at worst contrived. I could not fairly compare a mini falling over a cliff, and a ferrai towing a trailer up a steep road, and infer that "a mini has better acceleration than a ferrai" (even if it would have correct for that one case)
What I find most interesting is the similariaties between the emails you chose to post, and your initial benchmark "results". In the benchmarking you drew a general conclusion from one specific setup. In posting only the offensive email you have recieved, I would infer that you are trying to create a general impression that the only email you had recieved was offensive, which I find hard to believe. Drawing a general conclusion from a minority of bad cases, is one of the main problems many people had with your inital test.
You advertise as an independent testing/benchmarking company, and since you are doing these tests at your own expense, I assume you are trying to retain that image of giving fair and accurate results. If you are indeed interested in giving a balanced view, I would encourage you to post a representive sample of all the email you have recieved about these tests.
If, however, you are releasing the hate mail solely as an attempt at PR, I would remind you of a saying of Salvor Hardin (see Foundation) "An atom blaster can point both ways". By attempting to discredit the people who pointed out the flaws in your benchmark, and failing, the flaws become more and more apparent. The initial benchmark was unbalanced. The release of only selected emails was unbalanced. Is "unbalanced" the image an "impartial" testing company should be trying to have?
As a side point, IANAL, posting selected, offensive content, to try and give the impression that all linux user are immature/offensive, and to provoke a reaction from the less controled reader, could be categorised as "Flamebait", but there may be a risk that it could be also described as incitment to hatred. I assume you had different reasons for just posting the offensive emails.
There are two sides to every story, and if you wish for people to think you are a balanced company, you should give balanced views, and that includes showing a fair sample of the emails you have recieved. (And sent as a response, also, as we have no idea if these were a inital email, or as a response to something you said to them)
In short, may the FUD you spread be Fair, Unbiased, Data
>I thought that every distribution had to be downloadable?
No, it doesn't have to be. It all depends on the licencing involved. For example, if there was some commerical software on the distro, then that could be under a licence that you can't redistribute it.
Even if it was all under a GPL, that also doesn't require a download link. The GPL allows someone to sell software on CDROM only, it doesn't require them to have it available for download. However, the GPL does give people the right to request the full source (on similar distribution methods as the binary, IIRC, i.e you can't by the software on CD, and insist on having the source downloadable) and to re-dististribute the software they have bought for no charge, and to make it available for download themselves. Nowhere does the GPL say "you must make your product available for download via the net", all it says is that anyone may do so without the permission of the author
Most distros have an available download link, as it's in their best interests to have people going to them direct, as there is nothing legally stopping someone getting the GPL distro, and mirroring it elsewhere.
As an aside, this is a new distro which is based on Mandrake, which is based on Redhat, does this 2nd level basis cause more confusion or does it give a chance for the derived distros to fix the problems of the base distro?
>In middle of flame war, user in America posted an article to newsgroup, forging the From: address to be that of Mr Godfrey. >Mr Godfrey contacted his ISP, Demon, asking for them to remove the message from their servers >Demon (for a variety of reasons) said no. Godfrey sued for libel, and won.
What had always confused me, is why this was a libel case at all. (and why it was against demon, not the poster)
Consider a comparasion - I phone someone (and say I am Joe Soap) and proceed to express "bad" opinions about something or other, but not against Joe Soap himself Would I be done for libel? Would the phone company, if they refused to phone the reciever and apologise on behalf of Joe Soap? I don't think so, I think a fraud case against the forger would be more appropiate. Or was a libel case chosen as the laws are more liberal?
Also, if it was treated as fraud, demon could have done something about it from day one, as they are well within their AUP to cancel fraudelent postings (as it's abuse of the internet, as opposed to abuse on the internet, which their AUP allows them to take action on)
It would have died down a lot quicker and nicer all around, if the posting was cancelled in the grounds it was forged, and a simple explaination "someone forged my name, it wasn't me" was posted to the relevent newsgroup. It does make me wonder why this didn't happen. --
There are several good reasons I can think of of having more regular, but less dramitic, major version releases
Maturity: the kernel has setted down quite a lot at 2.2, and hopefully there won't be as much major upheavals as from 2.0 to 2.2, and it should me more an adding drivers/tweaking, rather than a redesign.
Marketing: or more accuratly, exposure. As the user market for linux gets more and more widespread, the impact of a major change will cause greater ripples. Most users (and distributions) will only run "stable" releases, so it makes sense to take lots of incremental upgrades (little steps) to that population, rather than on big jump (and possible fall over)
As the market of Linux increases, so does the potential interia to change. Take libc5-glibc for example, The linux bandwadgon hadn't really started rolling then, but if it had before the change, I think it would have been a lot more hassle to get people to migrate, because of all the apps that would have been released for libc5 systems.
As a result of this increasing inertia, I think companies will slowly drft from releaseing bleeding edge releases, "first with linux v x.y!!" to releasing "burnt-in" versons. For the end users, that's good, for the developers, that's bad, as there is less people to feedback bugs. Hence, the release stable versions of kernel often, will help allieviate this problem.
Doesn't this sound a little like the film "Mercury Rising", where a supersafe algorith is released as a simple puzzle, and a yough kid actually cracks it? --
You have pointed out one thing that has worried me also, that Redhat may have too much power over linux. I'm not talking power as in control, like MS has over windows, but power as in influnence, which RedHat have without doubt.
I would be very concerned that project may be cometing on marketing, not on merit. Even between two projects where Redhat has no interest, whichever one is distributed with RedHat gets more exposure, and how many "official rollouts" will specify "no third party (non-redhat signed) software) ?
Remember, infulence is like money in the bank, the less you use it, the more you have. With Redhat releasing everything (currently) under the GPL, and going public, that's a lot of influence piling up. What happens if they every try to cash in on that account?
>This is like suing the phone company because I said something about someone else I didn't like over a phone conversation.
It would be closer to a situation to where you were subjected to abusive phone calls, you complained to your phone company, and they refused to take any action on they grounds that they can't control what phone users say. And then you took a case against them for taking no action.
Personally, I think the initial action (that of complaining to Demon about a alleged libelous posting) was the wrong one to take, for two reasons (IANAL, etc.)
a) I don't think it was libelous, as AFAICS, the initial posting was forged to be from the poster, and expressed nasty opinions which were not about the author. This, I believe, would be more accurately classed as a "fraud", not a libel (any lawyers care to comment?)
b) Since the original posting was forged, the poster could have simply complained to demon about a forged posting, and Demon would have acted. A fraudlent posting is against their AUP, and I believe they could have cancelled the posting, and kicked out the offender, because their judgement would not be based on content. Their AUP is against abuse *of* the internet, not abuse *on* the internet.
Why the original poster didn't either cancel the posting himself, or get demon to do on his behalf (on the grounds it was forged, not libellious) and post an explaination on the same news group, I don't know. IMO, it would have come across a lot better to say "someone forged an posting from me, I have notified the ISP involved and action will be taken" rather than yell at the ISP "libel!" and go to the courts when they refuse (quite rightly too, IMO) to act on the grounds of content.
However, this is now a moot issue, and Demon have been found liable for an alleged libelous posting, and it does set a particular nasty precedent, that you can complain to an ISP about the content of a posting, and get the posting pulled. What happens I someone gets a posting of mine cancelled, and I complain to my ISP that the cancelling of that article causes me damage (eg I am replying contradicting stuff said about me) How should the ISP react? Cancel? Not cancel?
I think Demon are doing the best thing in the circumstance, by not taking it any further, as precedents could be set if the case went against them as in the lower courts.
I hope that the laws going through the court recognise the unique situations of ISP's, being somewhere between a common carrier, and a publisher. They are a common carrier, insofar as they shouldn't be held liable for what a customer says, but a publisher in the fact that unlike a phone company, what is say is stored, and propagated by them, after the fact.
It's a fine line between allowing the ISP to cancel/remove what it should be able to (spam, obvious illegal material) and forcing them to censor content and act on the complaint of one person.
>Funny thing is, if this were to happen in the US, the whole place would be bulldozed under >and replaced with the Bletchley Memorial Shopping Center
Or only US citizens would be allowed full access, other people would only be allowed access to 56 acres of the ground, and people from Iraq would be refused entry.:)
Seriously, I've been to Bletchley Park, and it's a interesting sight for those that are interested in crypo, or computers in general (They had a computer archive room when I was there, everything from 8" floppys, shells of PDP, ah memories:)
They also had a mockup of the original collusus, and thankfully they also had a lot of praise for the people that worked there, and didn't just say, "the computer did it". The sense of history was immense, seeing the actual rooms where the cracking was done, was awesome.
Although, my favourite part of the time I was there was when they had a flyover from a bomber, and a hurricane at about 200 ft - very impressive!:)
> If everyone paid per byte for their internet access there would be almost no lag, because people would be more careful about their usage.
Firstly, this would also make terribly complicated phone bills, would the ISP have to give you an itemised bill of where your bandwidth went? (60MB download from www.x.com, etc)
Also, when you talk about "metered" calls, what are you metering? Time? Bandwidth? A mixture of both? A long call takes up a modem slot, even if little traffic is downloaded.
So, you can: a) meter on time, (like BT local calls) b) unmetered with a higher flat rate. (leased line) c) meter on bandwidth (eg Tele2 ,
(With Tele2 you get a wireless link (arial goes into a ethernet card) You pay a flat charge a month, with an certain allowed bandwidth a month, charged for any overage)
Personally I prefer the last option, a full time line, with a reasonable bandwidth allowance. The starting tarrif is £60 a month, for a 128K line, and 150MB a month, which is tempting, but a little too expensive. The more people that get into this sort of pricing model, the more competive it will be, and the more the range of tarrifs will suit the consumer.
--
Re:Less mod upping, Default score.
on
Slashdot Tweaks
·
· Score: 1
AC comments are a difficult one, as they vary in quality so much. If I see a AC comment that's been scored up to 1, I'd read it, as it's a comment that someone thought was worth 1, rather than getting a 1 by default. However, a score of one, is still a score of one, even if IMO, an AC post of score 1 is usually worth more than logged in post of 1 For an AC comment to reach my overspill index threshold (+2), it has to be upped twice
Giving a poster who often gets moderated up an inital higher score is a good idea on the surface, but I think it's a bad thing, and there are always potential problems.
Remember the basic premise of the moderator guidelines: Concentrate on promotion, not demoting. With that in mind, if you give people with high averages high initial scores, not all of those posts will be good, and some will have to be marked down, wasting moderation points.
Another potential abuse if someone gets a high initial posting score, is to post a lot of comments to old threads, and suddenly they have 100 comments, all score 2, which will look even better to the system.
Also, even though under the idea you have there, I'd start on a 2, I don't want to! I want to earn that point for each comment I post. (I still remember the buzz when I got my first >1 post:) Compare that to feeling of the good poster, who gets "demoted" to a score of 1.
The difference in default posting between logged in and AC posts results in problems like in the first paragraph (good AC post =1 ; average logged in post =1), increasing the possible difference to 2, would exagerate this more.
Averaging as you laid it out, also favours people who post often ~2 , but doesn't favour those people who don't post very often, but post well.
As a point about moderation in general, there is a risk that it may lead to steotyping of views.
Ideally, moderators should be impartial to the opinion expressed, but they are only human. For example, if they see a badly worded comment, if they agree with it, it's natural to subconciously ignore the words, see the idea, and say "good post". If they don't agree with it, it will appear like nonsense. It's like seeing a pattern in ink blots, you can see randomness or patterns in the same blot.
So, Joe Soap, posts a comment and gets upped, he'll try and post something as good again, and maybe along the same opinion. And Joesph Soap, Esquire, a moderator, who sees a post he likes, and upps it. Fred Bloggs reads at +2, only sees those posts.
In short, there is no ideal scoring system, and this one has flaws, but it's quite good as it stands.
Welcome to your Microsoft Car. By opening the door of this car, you have agreed to this contact and are legally bound by its terms.
You are hearby granted licence to drive this car for purpose of leisure only, within the radius of 50 miles of point of purchase. Should you wish to use this care for commerical purposes, or for distances longer than 50 miles, you must purchase an upgrade to this licence, details of costs are available from your local MS office
This car comes without any warranty, evnt those assumed for fitness of purpose. If this car breaks down, we will disclaim liabilty, and not be liable for any damges resulting thereof.
This car has been fitted with the lastest cut-off system, whereby we can remotely imobilise your car in the case we are in suspicion of you breaching your licencening agreements (eg for commerical use, more than 50 miles, or listening to music in mp3 instead of microsoft format)
In no case will we be liable for the damages resulting from cutoff, not even if life or money is lost as a consequence therof. any fines for stopping in the midst of the freeway must be borne by the customer.
If you even wish to sell your car, you must contact us for permission to do so, permission may be granted in exceptional circumstances, the normal requiremnt is for a second owner to purchase the licence to use the car from us. We will retain any monies thereof, and you will also be unable to use the car hencewith.
You may not attempt to guess what is wrong with the car, if it splutters and stops, you may *not* assume that it is out of petrol and attempt to refuel. Yu must bring it to a MS approved dealer, where he will apply the required fix (for an appropiate fee). Attempting to refuel the car, will result in breach of the licence, and your car may be cut off at any day henceforth. The petrol gauge is for use of a Microsoft certificed mechanic for diagnositic purposes only.
As the car is the primary mode of transport, any other modes of transport are deemed copies, and are subject to patent laws. Any mode of device which transports a person, goods, information or thought from two distinct places will be covered. injunctions are currently in place against bicycles, wheelbarrows, televisions, and telepaths in the respective categories.
This casr is equipped with the latest map guidance, so it can tell you (and us) hwere exactly you are in the world. On each entry to the car, it will ask you "where do you want to go today?" If you answer correctly, it will transport you there. Any attempt to visit one of our competitiors will result in imediate terminition of your licence.
This MS care is fully compatible with all othe MS road users, howevrer any crashs as a result of contact with other road users will be deemed to be the fault of the non-MS road user, and MS will not be liable.
this product is only supported on MS stamdard road. For a definition of the word "supported" please see licence 345, section 4, paragraph 5, with excpemptions for cases detailed in sections 1 through 4.
Thank you for *choosing* an MSCAR, the only car that can get through an MS toll bridge in under 3 hours. Our competitors (which you are free to choose, of course, subject to fillout the relevent documentation) seem to be unable to cope with this simple transport protocol.
>That doesn't seem to be the mark of someone that won't accept input from others.
Actually, when I first heard about GNOME and Enlightentment, I went to the respective web sites to have a read. Gnome was okay, but on the enlightenment web site the following comments caught my eye:
"One thing you can be certain of is that it DOES work - the developers run it all day themselves - There is a reason it doesn't work for you - it is likely something on your system." and "Please make sure you also have a stable Xserver - Enlightenment can push X hard - often X breaks. When your server goes down - your server is at fault."
Those two sentences gave me the distinct *impression* that the authors didn't really care about the end users, "it works for us, if it doesn't work for you, it's your fault, not ours"
Of course, with any software, it is always possible to be a user configuration problem, and nothing to do with 'E', but the attitude that came across in those sentences would leave me very reluctant to even attempt to report a problem.
So before he throws too many stones about redhat believing users don't really count, he should check if he's living in a glass house.
This does seem to be solely based on the usage of a file format, and whether another author is allowed to use that file format to import into another program. (which of course is the *last* thing you want to happen to your file format!)
This seems groundless, as this sort of thing has been done for years, if you look through MS's site, it shows lots of white papers on how to migrate from lotus notes to exchange, for example, and excel can import various non MS formats.
However, if this stands, it could set a nasty precedent, imagine if gnumeric or koffice were not allowed to have an excel import option? Or if Samba was not able to use the SMB protocol?
The possible repercissions of "you're not allowed to build an import filter for our file format unless we allow it" do not bear thinking about.
Personally I believe either, and both could be equally correct in the correct context.
The GNU project was to create a free, open unix-like operating system, and as a result, there was a description of the target they were aiming for.
Linux is the correct name for the kernel, of that there is little doubt, but when people describe the whole system, that's when the fun starts.
If you consider GNU as an abstract description of a type (or characteristic) of OS (free/open unix clone), then saying Debian/Redhat etc are GNU systems is correct. Saying the same systems are Linux systems is equally correct, as that is the kernel they are based on.
The problem lies in that by saying GNU/linux, are you accrediting "ownership" (for want of a better term) of linux to GNU? I think not.
A few examples may make the point I'm trying to describe clearer
"The GNU compilier, gcc" is GNU in both abstract, and actual (cause they wrote it)
Redhat distribution is GNU/linux in the abstract term, but not in the actual
Debian is GNU/Linux is both, as that was Debian's choice to call it that.
So, Linux is acceptable to me, as is GNU/Linux, as the latter can refer to the abstraction (distribution has all the characteristic's of the GNU project) or the actual (this distribution was put togther based on GNU)
Does this make any sense, or is this all doublethink?
>But on CGI, you invoke a new process with each client request, no matter how many servers you've preforked or how many threads are idle. Presto: poor performance, no matter what the preforking parameters are.
Isn't this one of the things FastCgi is supposed to be fixing, instead of launching one process per perl script, it launched one perl interpeteror and passes it all the perl scripts, hence less overhead, and more speed (with the drawback that the scripts have to explicitly free memory and be slightly modified) (with a loop about the script)
Not quite thread like, but definitly not on process per cgi request.
Slashdot Mirror
Couple of points, some exploits don't work against a debug build system, that do in release mode - ever tried to track down a bug that was unreproducable when built in debug?
For another off the wall point, what if this is not actually running w2k, but Linux, and MS use the Anti-Ms brigade, to poke hole in Linux's default security?
One of the rules of engagement is that you have to tell them how you did it, but it doesn't give a time limit on when you have to tell them.
--
If reverse engineering is not allowed, would that now cause a problem for MicroSoft, since they reverse engineered the AIM client?
Seriously, without 100% open standards, outlawing reverse engineering is simple anticompetitive.
--
The CD-ROM ejecting on this machine...
Where you'd have to go to when people ask for your floppy...
going to a p0rn site and turning off the power switch by accident...
--
If you read the article, it says you get a month's equilavent of your original machines power.
/10757 = 4.01 AmdMinutes.
:)
so, a month of an average AMD350 is
30*24*60 = 43,200 AMDMinutes
which is worth
43,200
So, the proud donater of a 386SX25, could get of the magnitude of 1 second
--
>I've even seen resistance to Corporate LAN e-mail systems such as Exchange or Notes precisely because senders are authenticated.
>(For example, if a secretary sends a message from the Boss, the message will read From:Boss Sent By:Secretary = Boss gets mad because secretary can't impersonate him/her)
(Actually, in exchange they can do this, depending on the setting. You can either give the secretary "Send on Behalf of" privs which will do as you say or "Send As" privs, which will allow the secretary to completely act like the boss)
Speaking of Exchange, didn't MS say they were targeting ISP's with the next version of exchange (Platinum?)
Imagine the possible (worst case) timeline...
Phase (1) MS get a few major ISP's to use exchange as email backbone, supporting "legecy" SMTP support, but advanatges for intra exchange-enabled sites (user verification with NT Challege/response etc)
Phase (2) MS add more features which only work intra-exchange sites. ISP-Exchange clients now installed as default on all windows OS's
Phase (3) Problems sending/recieving to SMTP sites, and due to critical mass being achieved by ISP-exchange, (as all the windows pre-installed clients don't now work with SMTP) it is the SMTP sites that have to adapt.
Unlikely, yes, but MS would be one of the few companys that could have both the will, and the market clout to decommotise email protocols
SMTP may be old, and have weaknesses, but it is open. If we are not careful and plan an open upgrade path (past ESTMP, and to something better and different) someone else will.
--
From Men at Arms (I think)
--
please meet virtual machine A and B, who have been configured to do a reverse open relay check on RCPT, One a particular day, A wants to send some email to B...
A: "HELO, B"
B: "pleased to meet you. A"
A: "I have some mail for you"
B: (thinking, do I trust A? Let me see...)
(B starts working in the background)
B: HELO A
A: Pleased to meet you B
B: I have some email for you (hee hee, but I'm not going to send it, just checking your answer"
A: (thinking, do I trust B? let me see...)
(A starts working in ther background)
A: HELO B
B: Pleased to meet you again, A. Busy today, isn't it?
....
In short, it's a nice idea, but it would set up an endless loop.
--
Dear MindCraft,
Thank you for publishing some of the correspondence you have recieved from
some vocal linux users. What I find interesting that all of the email you have
displayed have been offensive and immature. While any reasonable person will
admit that in any mass of people there are extremes in any direction, the same
reasonable people would not infer the behaviour of the group form the actions
of a minority.
Your inital tests raised several issues (and found some bugs, now fixed -
thanks!) about the imbalance between the setup for NT and Linux, details of
which can be found on any reputable news source, so I shouldn't need to
reiterate them here. The general conclusion you attempted to dervive from that
specific setup (NT is x times faster than Linux) was at best flawed, at worst
contrived. I could not fairly compare a mini falling over a cliff, and a
ferrai towing a trailer up a steep road, and infer that "a mini has better
acceleration than a ferrai" (even if it would have correct for that one case)
What I find most interesting is the similariaties between the emails you chose
to post, and your initial benchmark "results". In the benchmarking you drew a
general conclusion from one specific setup. In posting only the offensive email
you have recieved, I would infer that you are trying to create a general
impression that the only email you had recieved was offensive, which I find
hard to believe. Drawing a general conclusion from a minority of bad cases, is
one of the main problems many people had with your inital test.
You advertise as an independent testing/benchmarking company, and since you
are doing these tests at your own expense, I assume you are trying to retain
that image of giving fair and accurate results. If you are indeed interested
in giving a balanced view, I would encourage you to post a representive
sample of all the email you have recieved about these tests.
If, however, you are releasing the hate mail solely as an attempt at PR, I
would remind you of a saying of Salvor Hardin (see Foundation) "An atom
blaster can point both ways". By attempting to discredit the people who
pointed out the flaws in your benchmark, and failing, the flaws become more
and more apparent. The initial benchmark was unbalanced. The release of only
selected emails was unbalanced. Is "unbalanced" the image an "impartial"
testing company should be trying to have?
As a side point, IANAL, posting selected, offensive content, to try and
give the impression that all linux user are immature/offensive, and to provoke
a reaction from the less controled reader, could be categorised as "Flamebait",
but there may be a risk that it could be also described as incitment to
hatred. I assume you had different reasons for just posting the offensive
emails.
There are two sides to every story, and if you wish for people to
think you are a balanced company, you should give balanced views, and that
includes showing a fair sample of the emails you have recieved. (And sent as a
response, also, as we have no idea if these were a inital email, or as a
response to something you said to them)
In short, may the FUD you spread be Fair, Unbiased, Data
Yours, etc.
--
Maybe we should start adding moderation comments to the articles posted to Slashdot.
I think this one would be worthy of a "-1 Troll/Flamebait"
--
>I thought that every distribution had to be downloadable?
No, it doesn't have to be. It all depends on the licencing involved. For example, if there was some commerical software on the distro, then that could be under a licence that you can't redistribute it.
Even if it was all under a GPL, that also doesn't require a download link. The GPL allows someone to sell software on CDROM only, it doesn't require them to have it available for download. However, the GPL does give people the right to request the full source (on similar distribution methods as the binary, IIRC, i.e you can't by the software on CD, and insist on having the source downloadable) and to re-dististribute the software they have bought for no charge, and to make it available for download themselves. Nowhere does the GPL say "you must make your product available for download via the net", all it says is that anyone may do so without the permission of the author
Most distros have an available download link, as it's in their best interests to have people going to them direct, as there is nothing legally stopping someone getting the GPL distro, and mirroring it elsewhere.
As an aside, this is a new distro which is based on Mandrake, which is based on Redhat, does this 2nd level basis cause more confusion or does it give a chance for the derived distros to fix the problems of the base distro?
--
>In middle of flame war, user in America posted an article to newsgroup, forging the From: address to be that of Mr Godfrey.
>Mr Godfrey contacted his ISP, Demon, asking for them to remove the message from their servers
>Demon (for a variety of reasons) said no. Godfrey sued for libel, and won.
What had always confused me, is why this was a libel case at all. (and why it was against demon, not the poster)
Consider a comparasion - I phone someone (and say I am Joe Soap) and proceed to express "bad" opinions about something or other, but not against Joe Soap himself Would I be done for libel? Would the phone company, if they refused to phone the reciever and apologise on behalf of Joe Soap? I don't think so, I think a fraud case against the forger would be more appropiate. Or was a libel case chosen as the laws are more liberal?
Also, if it was treated as fraud, demon could have done something about it from day one, as they are well within their AUP to cancel fraudelent postings (as it's abuse of the internet, as opposed to abuse on the internet, which their AUP allows them to take action on)
It would have died down a lot quicker and nicer all around, if the posting was cancelled in the grounds it was forged, and a simple explaination "someone forged my name, it wasn't me" was posted to the relevent newsgroup. It does make me wonder why this didn't happen.
--
There are several good reasons I can think of of having more regular, but less dramitic, major version releases
Maturity: the kernel has setted down quite a lot at 2.2, and hopefully there won't be as much major upheavals as from 2.0 to 2.2, and it should me more an adding drivers/tweaking, rather than a redesign.
Marketing: or more accuratly, exposure. As the user market for linux gets more and more widespread, the impact of a major change will cause greater ripples. Most users (and distributions) will only run "stable" releases, so it makes sense to take lots of incremental upgrades (little steps) to that population, rather than on big jump (and possible fall over)
As the market of Linux increases, so does the potential interia to change. Take libc5-glibc for example, The linux bandwadgon hadn't really started rolling then, but if it had before the change, I think it would have been a lot more hassle to get people to migrate, because of all the apps that would have been released for libc5 systems.
As a result of this increasing inertia, I think companies will slowly drft from releaseing bleeding edge releases, "first with linux v x.y!!" to releasing "burnt-in" versons. For the end users, that's good, for the developers, that's bad, as there is less people to feedback bugs. Hence, the release stable versions of kernel often, will help allieviate this problem.
--
Doesn't this sound a little like the film "Mercury Rising", where a supersafe algorith is released as a simple puzzle, and a yough kid actually cracks it?
--
Say that too fast, and it sounds like "Stalin" :)
--
S/WAN is another reason too, it's designed for the 2.0 series kernel, and is unlikely to work with the 2.2 series for a while yet.
In short though, the point with linux kernels is usually "do I *need* to upgrade?" rather than "why *shouldn't* I upgrade?"
--
You have pointed out one thing that has worried me also, that Redhat may have too much power over linux. I'm not talking power as in control, like MS has over windows, but power as in influnence, which RedHat have without doubt.
I would be very concerned that project may be cometing on marketing, not on merit. Even between two projects where Redhat has no interest, whichever one is distributed with RedHat gets more exposure, and how many "official rollouts" will specify "no third party (non-redhat signed) software) ?
Remember, infulence is like money in the bank, the less you use it, the more you have. With Redhat releasing everything (currently) under the GPL, and going public, that's a lot of influence piling up. What happens if they every try to cash in on that account?
--
>This is like suing the phone company because I said something about someone else I didn't like over a phone conversation.
It would be closer to a situation to where you were subjected to abusive phone calls, you complained to your phone company, and they refused to take any action on they grounds that they can't control what phone users say. And then you took a case against them for taking no action.
Personally, I think the initial action (that of complaining to Demon about a alleged libelous posting) was the wrong one to take, for two reasons (IANAL, etc.)
a) I don't think it was libelous, as AFAICS, the initial posting was forged to be from the poster, and expressed nasty opinions which were not about the author. This, I believe, would be more accurately classed as a "fraud", not a libel (any lawyers care to comment?)
b) Since the original posting was forged, the poster could have simply complained to demon about a forged posting, and Demon would have acted. A fraudlent posting is against their AUP, and I believe they could have cancelled the posting, and kicked out the offender, because their judgement would not be based on content. Their AUP is against abuse *of* the internet, not abuse *on* the internet.
Why the original poster didn't either cancel the posting himself, or get demon to do on his behalf (on the grounds it was forged, not libellious) and post an explaination on the same news group, I don't know. IMO, it would have come across a lot better to say "someone forged an posting from me, I have notified the ISP involved and action will be taken" rather than yell at the ISP "libel!" and go to the courts when they refuse (quite rightly too, IMO) to act on the grounds of content.
However, this is now a moot issue, and Demon have been found liable for an alleged libelous posting, and it does set a particular nasty precedent, that you can complain to an ISP about the content of a posting, and get the posting pulled. What happens I someone gets a posting of mine cancelled, and I complain to my ISP that the cancelling of that article causes me damage (eg I am replying contradicting stuff said about me) How should the ISP react? Cancel? Not cancel?
I think Demon are doing the best thing in the circumstance, by not taking it any further, as precedents could be set if the case went against them as in the lower courts.
I hope that the laws going through the court recognise the unique situations of ISP's, being somewhere between a common carrier, and a publisher. They are a common carrier, insofar as they shouldn't be held liable for what a customer says, but a publisher in the fact that unlike a phone company, what is say is stored, and propagated by them, after the fact.
It's a fine line between allowing the ISP to cancel/remove what it should be able to (spam, obvious illegal material) and forcing them to censor content and act on the complaint of one person.
--
>Funny thing is, if this were to happen in the US, the whole place would be bulldozed under
:)
:)
:)
>and replaced with the Bletchley Memorial Shopping Center
Or only US citizens would be allowed full access, other people would only be allowed access to 56 acres of the ground, and people from Iraq would be refused entry.
Seriously, I've been to Bletchley Park, and it's a interesting sight for those that are interested in crypo, or computers in general (They had a computer archive room when I was there, everything from 8" floppys, shells of PDP, ah memories
They also had a mockup of the original collusus, and thankfully they also had a lot of praise for the people that worked there, and didn't just say, "the computer did it". The sense of history was immense, seeing the actual rooms where the cracking was done, was awesome.
Although, my favourite part of the time I was there was when they had a flyover from a bomber, and a hurricane at about 200 ft - very impressive!
--
> If everyone paid per byte for their internet access there would be almost no lag, because people would be more careful about their usage.
Firstly, this would also make terribly complicated phone bills, would the ISP have to give you an itemised bill of where your bandwidth went? (60MB download from www.x.com, etc)
Also, when you talk about "metered" calls, what are you metering? Time? Bandwidth? A mixture of both? A long call takes up a modem slot, even if little traffic is downloaded.
So, you can:
a) meter on time, (like BT local calls)
b) unmetered with a higher flat rate. (leased line)
c) meter on bandwidth (eg Tele2 ,
(With Tele2 you get a wireless link (arial goes into a ethernet card) You pay a flat charge a month, with an certain allowed bandwidth a month, charged for any overage)
Personally I prefer the last option, a full time line, with a reasonable bandwidth allowance.
The starting tarrif is £60 a month, for a 128K line, and 150MB a month, which is tempting, but a little too expensive. The more people that get into this sort of pricing model, the more competive it will be, and the more the range of tarrifs will suit the consumer.
--
AC comments are a difficult one, as they vary in quality so much. If I see a AC comment that's been scored up to 1, I'd read it, as it's a comment that someone thought was worth 1, rather than getting a 1 by default. However, a score of one, is still a score of one, even if IMO, an AC post of score 1 is usually worth more than logged in post of 1
:) Compare that to feeling of the good poster, who gets "demoted" to a score of 1.
For an AC comment to reach my overspill index threshold (+2), it has to be upped twice
Giving a poster who often gets moderated up an inital higher score is a good idea on the surface, but I think it's a bad thing, and there are always potential problems.
Remember the basic premise of the moderator guidelines: Concentrate on promotion, not demoting. With that in mind, if you give people with high averages high initial scores, not all of those posts will be good, and some will have to be marked down, wasting moderation points.
Another potential abuse if someone gets a high initial posting score, is to post a lot of comments to old threads, and suddenly they have 100 comments, all score 2, which will look even better to the system.
Also, even though under the idea you have there, I'd start on a 2, I don't want to! I want to earn that point for each comment I post. (I still remember the buzz when I got my first >1 post
The difference in default posting between logged in and AC posts results in problems like in the first paragraph (good AC post =1 ; average logged in post =1), increasing the possible difference to 2, would exagerate this more.
Averaging as you laid it out, also favours people who post often ~2 , but doesn't favour those people who don't post very often, but post well.
As a point about moderation in general, there is a risk that it may lead to steotyping of views.
Ideally, moderators should be impartial to the opinion expressed, but they are only human. For example, if they see a badly worded comment, if they agree with it, it's natural to subconciously ignore the words, see the idea, and say "good post". If they don't agree with it, it will appear like nonsense. It's like seeing a pattern in ink blots, you can see randomness or patterns in the same blot.
So, Joe Soap, posts a comment and gets upped, he'll try and post something as good again, and maybe along the same opinion. And Joesph Soap, Esquire, a moderator, who sees a post he likes, and upps it. Fred Bloggs reads at +2, only sees those posts.
In short, there is no ideal scoring system, and this one has flaws, but it's quite good as it stands.
--
(licence contained on dashboard)
Welcome to your Microsoft Car. By opening the door of this car, you have agreed to this contact and are legally bound by its terms.
You are hearby granted licence to drive this car for purpose of leisure only, within the radius of 50 miles of point of purchase.
Should you wish to use this care for commerical purposes, or for distances longer than 50 miles, you must purchase an upgrade to this licence, details of costs are available from your local MS office
This car comes without any warranty, evnt those assumed for fitness of purpose.
If this car breaks down, we will disclaim liabilty, and not be liable for any damges resulting thereof.
This car has been fitted with the lastest cut-off system, whereby we can remotely imobilise your car in the case we are in suspicion of you breaching your licencening agreements (eg for commerical use, more than 50 miles, or listening to music in mp3 instead of microsoft format)
In no case will we be liable for the damages resulting from cutoff, not even if life or money is lost as a consequence therof. any fines for stopping in the midst of the freeway must be borne by the customer.
If you even wish to sell your car, you must contact us for permission to do so, permission may be granted in exceptional circumstances, the normal requiremnt is for a second owner to purchase the licence to use the car from us. We will retain any monies thereof, and you will also be unable to use the car hencewith.
You may not attempt to guess what is wrong with the car, if it splutters and stops, you may *not* assume that it is out of petrol and attempt to refuel. Yu must bring it to a MS approved dealer, where he will apply the required fix (for an appropiate fee). Attempting to refuel the car, will result in breach of the licence, and your car may be cut off at any day henceforth. The petrol gauge is for use of a Microsoft certificed mechanic for diagnositic purposes only.
As the car is the primary mode of transport, any other modes of transport are deemed copies, and are subject to patent laws. Any mode of device which transports a person, goods, information or thought from two distinct places will be covered. injunctions are currently in place against bicycles, wheelbarrows, televisions, and telepaths in the respective categories.
This casr is equipped with the latest map guidance, so it can tell you (and us) hwere exactly you are in the world. On each entry to the car, it will ask you "where do you want to go today?" If you answer correctly, it will transport you there. Any attempt to visit one of our competitiors will result in imediate terminition of your licence.
This MS care is fully compatible with all othe MS road users, howevrer any crashs as a result of contact with other road users will be deemed to be the fault of the non-MS road user, and MS will not be liable.
this product is only supported on MS stamdard road. For a definition of the word "supported" please see licence 345, section 4, paragraph 5, with excpemptions for cases detailed in sections 1 through 4.
Thank you for *choosing* an MSCAR, the only car that can get through an MS toll bridge in under 3 hours. Our competitors (which you are free to choose, of course, subject to fillout the relevent documentation) seem to be unable to cope with this simple transport protocol.
--
>That doesn't seem to be the mark of someone that won't accept input from others.
Actually, when I first heard about GNOME and Enlightentment, I went to the respective web sites to have a read. Gnome was okay, but on the enlightenment web site the following comments caught my eye:
"One thing you can be certain of is that it DOES work - the developers run it all day themselves - There is a reason it doesn't work for you - it is likely something on your system."
and
"Please make sure you also have a stable Xserver - Enlightenment can push X hard - often X breaks. When your server goes down - your server is at fault."
Those two sentences gave me the distinct *impression* that the authors didn't really care about the end users, "it works for us, if it doesn't work for you, it's your fault, not ours"
Of course, with any software, it is always possible to be a user configuration problem, and nothing to do with 'E', but the attitude that came across in those sentences would leave me very reluctant to even attempt to report a problem.
So before he throws too many stones about redhat believing users don't really count, he should check if he's living in a glass house.
--
This does seem to be solely based on the usage of a file format, and whether another author is allowed to use that file format to import into another program. (which of course is the *last* thing you want to happen to your file format!)
This seems groundless, as this sort of thing has been done for years, if you look through MS's site, it shows lots of white papers on how to migrate from lotus notes to exchange, for example, and excel can import various non MS formats.
However, if this stands, it could set a nasty precedent, imagine if gnumeric or koffice were not allowed to have an excel import option? Or if Samba was not able to use the SMB protocol?
The possible repercissions of "you're not allowed to build an import filter for our file format unless we allow it" do not bear thinking about.
--
Personally I believe either, and both could be equally correct in the correct context.
The GNU project was to create a free, open unix-like operating system, and as a result, there was a description of the target they were aiming for.
Linux is the correct name for the kernel, of that there is little doubt, but when people describe the whole system, that's when the fun starts.
If you consider GNU as an abstract description of a type (or characteristic) of OS (free/open unix clone), then saying Debian/Redhat etc are GNU systems is correct. Saying the same systems are Linux systems is equally correct, as that is the kernel they are based on.
The problem lies in that by saying GNU/linux, are you accrediting "ownership" (for want of a better term) of linux to GNU? I think not.
A few examples may make the point I'm trying to describe clearer
"The GNU compilier, gcc" is GNU in both abstract, and actual (cause they wrote it)
Redhat distribution is GNU/linux in the abstract term, but not in the actual
Debian is GNU/Linux is both, as that was Debian's choice to call it that.
So, Linux is acceptable to me, as is GNU/Linux, as the latter can refer to the abstraction (distribution has all the characteristic's of the GNU project) or the actual (this distribution was put togther based on GNU)
Does this make any sense, or is this all doublethink?
--
Isn't this one of the things FastCgi is supposed to be fixing, instead of launching one process per perl script, it launched one perl interpeteror and passes it all the perl scripts, hence less overhead, and more speed (with the drawback that the scripts have to explicitly free memory and be slightly modified) (with a loop about the script)
Not quite thread like, but definitly not on process per cgi request.
--