People still get spam? Haha, they're probably still tormented by popups and limited to browsing in a single tab!
Seriously, I almost never have a spammer get through to my inbox, because I have spent a little time up front in prevention: 1) Switch off that preview pane for HTML messages. (Every time you look at one of those things, it is a confirmed hit for the spammer). 2) Install a bayesian filter on your local machine (I use PopFile as a classification tool, but if you just want to prevent spam, then Thunderbird's builtin seems ok too) 3) Train bayesian filter 4) Once per week, scan the email subjects of your "spam folder" for items that may have been mis-classified. Repeat (4) as needed. 5) Never post on a newsgroup with your real email address. Ever. You will regret it. Some viruses/worms scan newsgroups for email addresses, and then send themselves to you. Even with spam prevention, it still hogs your email bandwidth and space usage.
Realistically, most people will do none of the above. The whole spam problem can only really be solved with new email clients, which do all of the above automatically.
It will never be solved legistatively, because there are always special interests that cause loopholes. And technical changes just impose a higher barrier to entry; they do not prevent the problem, just consolidate it to a few powerful spam-service providers.
Re:source code escrow not very useful
on
Source Code Escrow
·
· Score: 1
On your point (b), I think that IE is currently almost unmaintainable, hence the new month-long turnaround on patches. Remember, this is a product that has been evolving with the WWW, so it has been extended in ways that the original designers never dreamed of.
So, there will never be a patched version of IE6 that doesn't suck. I have heard that the next version of IE is a re-write though, so code quality should be better then.
Its all moot though; we will never see the source code of anything MS unless there is a clear monetary gain to be made, or open source becomes wildly popular.
I think that threaded email is only half of the solution. Some of my conversations use email, some Usenet, some use instant messaging software, some use issue tracking software, some use phone calls, and the rest are person-person.
Threaded emails is nice, but really it would be great if I had threaded multi-provider tracking of conversations. So, if a IM conversation leads to an email + a phone call, it would be great if that could all be captured in a threaded view.
Its all technically feasible, except for (perhaps) the person-person chats.
I have a DirecTivo, and am very happy, except for one thing - it will not dialup to DirectTV over Vonage (IP Telephone). The TiVo dialup works fine, but the DirectTV tries to connect at too high a speed. And it is not as "hackable" as other Tivo units, so it is nigh impossible to use USB ethernet instead of dialup.
Otherwise, its just dandy. Dual tuners, so I can record 2 things, or watch one and record another, etc. I think it costs me around $5 per month.
Outsourcing works fine for manufacturing, and probably a bunch of other service oriented things. The common aspect of the things it works for, are that the problems are well defined. Wind it up and release it, and awaayyy you go.
For programming, outsourcing is a spectacularly bad idea, because it is almost never well defined. In fact, in my experience, the quality of a programming product is driven by constant communication between the lead programmers and those that want the software developed.
I just cannot see how that can consistently happen when outsourcing to a different location, in a different timezone, with different accents and native languages and no face-to-face communication.
Your comments regarding the downsides of your online system are not indicative of all online backup systems. You paid nothing for your online backup software, and got what you paid for in terms of features.
A good online backup system will keep many months/years of data in an easily accessible manner, because when you are only ever storing the delta data, storage is very efficient.
I took a look at easeBackup (http://www.kiesoft.com/) and it looks good from a UI point of view. However, I would have some concerns, especially if using it for offsite backup.
I would not trust the security of data unless the vendor published their encryption and security standards and algorithms.
They state they use 2048 bit PGP encryption, which says nothing really, because they do not state how it is used.
The fact that you are backing up 40Gigs per night speaks more to the inefficiency of the backup method than the actual size of data. I doubt your data actually grows at that rate. (40G * 365 = 1.5TB per year?)
The point being, online backups are more bandwidth efficient, because they have to be. So, using online backup, your 40Gigs per night may drop to 1G, or even less.
Regarding security, there are ways to backup data so that the data itself is kept completely encrypted on the remote site, in such a way that even the backup provider cannot unencrypt it. Safe from packet sniffers, as well as the RIAA or other big brothers.
Disks have other benefits * they are random access, so incremental backups can easily merge into a complete backup. * they are faster than tape, both for restoring and backing up data.
You're overlooking the random salt value, which is different on each machine, or even the same machine for different accounts. So, generally you would have a hash of "$salt-$password".
The problem mentioned with crypt() is that the $salt value is very small (1.5 bytes), thus making dictionary attacks feasible. If MySQL is not using salt values, then they risk similar attacks.
I had a similar problem on my wife's desktop yesterday. A little googling suggested the solution was to switch the XP CD-writing capability off, then on again. I did this in the properties dialog of the drive (told XP that it was not a CD burner, applied the change, then told it that it was a CD burner), and it worked fine after that.
Now if only I could get my XP Pro box to Standby without hanging, then I'd be a happy clam.
Its been said before, but I'll say it again...Americans do not have a monopoly on IT skills and experience.
As for MIT graduates, education is a helping hand, not a substitute for years of software development experience. If I want business critical software developed, then I want to give the job to someone who has a proven track record.
I'm not saying anything about the specific situation described in the article - I cannot speak to the experience of the Indian programmers, or the reasons for hiring them. For all I know, they were inexperienced and produced a buggy, unusable, unmaintainable product. But I doubt that students would have done any better.
The license agreement states it is "LIMITED TIME FREE SOFTWARE" (or something like that). They have left the door open to charging at a later stage.
The rest of the license agreement is somewhat hard to read, due to a lack of headings and an abundance of lawyer-speak.
I'm not sure that their business model makes much sense though - I expect that at this stage of the game it will be too hard to grab a good share of the IM market, especially with only the one distinguishing feature, and no message encryption.
ASP.NET is definitely ahead of vanilla ASP. It has separated code from display, which is good. It has also added web controls, which are very very powerful. It is also (mostly) compiled, which adds speed.
However, it is still lacking in several areas: * lack of built-in page templating mechanism * datagrids are latebound (and possibly interpreted) * CSS support in Visual Studio is abysmal * the style of ASP.NET is "lots of custom pages" - it does not encourage large maintainable web sites.
All in all, it is very "VB" - nice and easy to create something simple, but it takes some ingenuity to create a larger, maintainable product.
.NET is very Windows-UNspecific, not at all what you would expect from MS. It has some Microsoft specific stuff, (eg. accessing windows registry) but that is generally relegated to a separate "Microsoft" namespace.
In addition, you seem to think that.NET is all about Windows Forms. It is not. For me, that is the least interesting aspect. You can write console applications, or server deamons, or web services, or DLLs for re-use by multiple UIs.
Like Java, C# is here NOW, and C# developers are here now. As are VB.NET developers, as fast as they can convince their management to change. As a platform,.NET has a large amount of support, from the people who write software for Microsoft platforms. I assure you, there are very many.
For myself, I develop specifically for Windows, and.NET has to be very important to my career. And Mono is too, because it opens up a whole new market for my MS-centric skills - suddenly, I have a chance of telling management that they are no longer tied to MS because their business relies on my Windows program. And that is very exciting.
If you are a Java specialist, there is really no need to stray, for now. However, Microsoft will do everything it can to make it easier and easier for you to migrate your skills to.NET, and at some point in the future it may become worth your while.
All of the arguments about I/O, porting code etc are pointless, because it eventually comes down to economics. If a company is spending $1,000,000 per year maintaining a mainframe, with code that can no longer grow, then there will come a time when they can no longer be competitive - they will have a huge overhead compared to their competitors, and will not be able to respond to market changes.
Mainframes *will* go away, *except* where there is not enough competition, or the cost of the mainframe is small in comparison with other cost centers.
I think that big banks will keep their mainframes, but eventually smaller, more agile banks will grow enough that the bigger banks can no longer compete. Then the big bank will either migrate or die.
Um. Let me think. I'm a member of a panel and I have consented to let big brother monitor my PC. Big brother starts suing people who illegally download music. Do I...
A) Don't worry, they're only after 12 year old girls B) Delete all evidence so they cannot sue me
Given the above, are my actions representative of other people who have *not* consented to being monitored? Clearly, no - therefore this falls into the 78.34% of statistics that are statistically meaningless.
Two weeks ago, I would probably have agreed with you. But I have installed various Linux flavors recently, and some of the recent distributions are really very good, even for a linux newbie like me. I do not think there is very far to go before at least one of the distributions hits the usability sweet spot.
Earlier this week I tried Knoppix (Debian based): Technically, I have not been so wowed in quite some time. I understand there are projects that will make it possible for a newbie to boot using Knoppix, check out whether he likes what he sees, then install it to his hard drive! All it needs is some market push, (how expensive is it to flood the market with CDs, like AOL does?), some nice tools to automagically migrate outlook express settings and email, and bam, a big chunk of desktop MS market share could easily go out of the Windows(TM).
Then there is Damn Small Linux, which fits on one of those credit card sized CDs. Not for your average Windows newbie, but wow, it just shows what you can do when you have such a good solid base to work off of.
In the end, Linux is a technically superior product that is only a little bit behind wrt usability - and the gap is closing very fast.
I am not sure that this still holds true. Perhaps it held true when the world was much more isolated, but I cannot think of any recent instances where the world suddenly iconized someone long since dead.
I think if Shakespeare was alive today we would all know about him and be enjoying his movies, and he would be very happy in his Hollywood mansion.
I disagree with the implication that it was harder for people in more ancient times to set aside the time and effort to be inventive. I would assert that the opposite is true...
I mean, people used to have very hard-working days, but in their evenings, they probably had much more time to themselves than we do today. TV and computer games are the big culprits in today's world, stealing all of our free time and numbing our minds.
Its like another recent slashdot discussion about how abundance of something previously available in limited supply often causes unexpected results. We have an overabundance of entertainment, and we feast on it because we cannot help ourselves, thus sacrificing other more wholesome and productive pursuits.
I have not RTFA, but it makes sense to be that the spam response rates are high for the people that took the survey, simply because the *same people* that respond to meaningless surveys are the ones that respond to spam.
In other words, the survey technique has skewed the results, and is (like 99.81% of statistics), meaningless.
Slashdot Mirror
People still get spam? Haha, they're probably still tormented by popups and limited to browsing in a single tab!
Seriously, I almost never have a spammer get through to my inbox, because I have spent a little time up front in prevention:
1) Switch off that preview pane for HTML messages. (Every time you look at one of those things, it is a confirmed hit for the spammer).
2) Install a bayesian filter on your local machine (I use PopFile as a classification tool, but if you just want to prevent spam, then Thunderbird's builtin seems ok too)
3) Train bayesian filter
4) Once per week, scan the email subjects of your "spam folder" for items that may have been mis-classified. Repeat (4) as needed.
5) Never post on a newsgroup with your real email address. Ever. You will regret it. Some viruses/worms scan newsgroups for email addresses, and then send themselves to you. Even with spam prevention, it still hogs your email bandwidth and space usage.
Realistically, most people will do none of the above. The whole spam problem can only really be solved with new email clients, which do all of the above automatically.
It will never be solved legistatively, because there are always special interests that cause loopholes. And technical changes just impose a higher barrier to entry; they do not prevent the problem, just consolidate it to a few powerful spam-service providers.
On your point (b), I think that IE is currently almost unmaintainable, hence the new month-long turnaround on patches. Remember, this is a product that has been evolving with the WWW, so it has been extended in ways that the original designers never dreamed of.
So, there will never be a patched version of IE6 that doesn't suck. I have heard that the next version of IE is a re-write though, so code quality should be better then.
Its all moot though; we will never see the source code of anything MS unless there is a clear monetary gain to be made, or open source becomes wildly popular.
I think that threaded email is only half of the solution. Some of my conversations use email, some Usenet, some use instant messaging software, some use issue tracking software, some use phone calls, and the rest are person-person.
Threaded emails is nice, but really it would be great if I had threaded multi-provider tracking of conversations. So, if a IM conversation leads to an email + a phone call, it would be great if that could all be captured in a threaded view.
Its all technically feasible, except for (perhaps) the person-person chats.
I can see the headlines now:
:/
"DirectTv sues all owners of ROM Burners".
It could happen
I have a DirecTivo, and am very happy, except for one thing - it will not dialup to DirectTV over Vonage (IP Telephone). The TiVo dialup works fine, but the DirectTV tries to connect at too high a speed. And it is not as "hackable" as other Tivo units, so it is nigh impossible to use USB ethernet instead of dialup.
Otherwise, its just dandy. Dual tuners, so I can record 2 things, or watch one and record another, etc. I think it costs me around $5 per month.
Outsourcing works fine for manufacturing, and probably a bunch of other service oriented things. The common aspect of the things it works for, are that the problems are well defined. Wind it up and release it, and awaayyy you go.
For programming, outsourcing is a spectacularly bad idea, because it is almost never well defined. In fact, in my experience, the quality of a programming product is driven by constant communication between the lead programmers and those that want the software developed.
I just cannot see how that can consistently happen when outsourcing to a different location, in a different timezone, with different accents and native languages and no face-to-face communication.
Your comments regarding the downsides of your online system are not indicative of all online backup systems. You paid nothing for your online backup software, and got what you paid for in terms of features.
A good online backup system will keep many months/years of data in an easily accessible manner, because when you are only ever storing the delta data, storage is very efficient.
I took a look at easeBackup (http://www.kiesoft.com/) and it looks good from a UI point of view. However, I would have some concerns, especially if using it for offsite backup.
I would not trust the security of data unless the vendor published their encryption and security standards and algorithms.
They state they use 2048 bit PGP encryption, which says nothing really, because they do not state how it is used.
The fact that you are backing up 40Gigs per night speaks more to the inefficiency of the backup method than the actual size of data. I doubt your data actually grows at that rate. (40G * 365 = 1.5TB per year?)
The point being, online backups are more bandwidth efficient, because they have to be. So, using online backup, your 40Gigs per night may drop to 1G, or even less.
Regarding security, there are ways to backup data so that the data itself is kept completely encrypted on the remote site, in such a way that even the backup provider cannot unencrypt it. Safe from packet sniffers, as well as the RIAA or other big brothers.
Is that 4 cents or 40 cents per tape GB?
Disks have other benefits
* they are random access, so incremental backups can easily merge into a complete backup.
* they are faster than tape, both for restoring and backing up data.
You're overlooking the random salt value, which is different on each machine, or even the same machine for different accounts. So, generally you would have a hash of "$salt-$password".
The problem mentioned with crypt() is that the $salt value is very small (1.5 bytes), thus making dictionary attacks feasible. If MySQL is not using salt values, then they risk similar attacks.
I had a similar problem on my wife's desktop yesterday. A little googling suggested the solution was to switch the XP CD-writing capability off, then on again. I did this in the properties dialog of the drive (told XP that it was not a CD burner, applied the change, then told it that it was a CD burner), and it worked fine after that.
Now if only I could get my XP Pro box to Standby without hanging, then I'd be a happy clam.
Its been said before, but I'll say it again...Americans do not have a monopoly on IT skills and experience.
As for MIT graduates, education is a helping hand, not a substitute for years of software development experience. If I want business critical software developed, then I want to give the job to someone who has a proven track record.
I'm not saying anything about the specific situation described in the article - I cannot speak to the experience of the Indian programmers, or the reasons for hiring them. For all I know, they were inexperienced and produced a buggy, unusable, unmaintainable product. But I doubt that students would have done any better.
I don't see what the big deal is. It would be foolish to get IT students to write business critical software, since they do not have the experience.
Let them graduate, and work under an experienced team lead, and then, IF they can make it cheaper than I can buy it, they can have the job.
The license agreement states it is "LIMITED TIME FREE SOFTWARE" (or something like that). They have left the door open to charging at a later stage.
The rest of the license agreement is somewhat hard to read, due to a lack of headings and an abundance of lawyer-speak.
I'm not sure that their business model makes much sense though - I expect that at this stage of the game it will be too hard to grab a good share of the IM market, especially with only the one distinguishing feature, and no message encryption.
ASP.NET is definitely ahead of vanilla ASP. It has separated code from display, which is good. It has also added web controls, which are very very powerful. It is also (mostly) compiled, which adds speed.
However, it is still lacking in several areas:
* lack of built-in page templating mechanism
* datagrids are latebound (and possibly interpreted)
* CSS support in Visual Studio is abysmal
* the style of ASP.NET is "lots of custom pages" - it does not encourage large maintainable web sites.
All in all, it is very "VB" - nice and easy to create something simple, but it takes some ingenuity to create a larger, maintainable product.
.NET is very Windows-UNspecific, not at all what you would expect from MS. It has some Microsoft specific stuff, (eg. accessing windows registry) but that is generally relegated to a separate "Microsoft" namespace.
.NET is all about Windows Forms. It is not. For me, that is the least interesting aspect. You can write console applications, or server deamons, or web services, or DLLs for re-use by multiple UIs.
.NET has a large amount of support, from the people who write software for Microsoft platforms. I assure you, there are very many.
.NET has to be very important to my career. And Mono is too, because it opens up a whole new market for my MS-centric skills - suddenly, I have a chance of telling management that they are no longer tied to MS because their business relies on my Windows program. And that is very exciting.
.NET, and at some point in the future it may become worth your while.
In addition, you seem to think that
Like Java, C# is here NOW, and C# developers are here now. As are VB.NET developers, as fast as they can convince their management to change. As a platform,
For myself, I develop specifically for Windows, and
If you are a Java specialist, there is really no need to stray, for now. However, Microsoft will do everything it can to make it easier and easier for you to migrate your skills to
Thats essentially the way BitTorrent works.
Its an evil plot to map slashdot users. He's probably logging all the IPs of the people who hit his website today!
All of the arguments about I/O, porting code etc are pointless, because it eventually comes down to economics. If a company is spending $1,000,000 per year maintaining a mainframe, with code that can no longer grow, then there will come a time when they can no longer be competitive - they will have a huge overhead compared to their competitors, and will not be able to respond to market changes.
Mainframes *will* go away, *except* where there is not enough competition, or the cost of the mainframe is small in comparison with other cost centers.
I think that big banks will keep their mainframes, but eventually smaller, more agile banks will grow enough that the bigger banks can no longer compete. Then the big bank will either migrate or die.
Um. Let me think. I'm a member of a panel and I have consented to let big brother monitor my PC. Big brother starts suing people who illegally download music. Do I...
A) Don't worry, they're only after 12 year old girls
B) Delete all evidence so they cannot sue me
Given the above, are my actions representative of other people who have *not* consented to being monitored? Clearly, no - therefore this falls into the 78.34% of statistics that are statistically meaningless.
Two weeks ago, I would probably have agreed with you. But I have installed various Linux flavors recently, and some of the recent distributions are really very good, even for a linux newbie like me. I do not think there is very far to go before at least one of the distributions hits the usability sweet spot.
Earlier this week I tried Knoppix (Debian based): Technically, I have not been so wowed in quite some time. I understand there are projects that will make it possible for a newbie to boot using Knoppix, check out whether he likes what he sees, then install it to his hard drive! All it needs is some market push, (how expensive is it to flood the market with CDs, like AOL does?), some nice tools to automagically migrate outlook express settings and email, and bam, a big chunk of desktop MS market share could easily go out of the Windows(TM).
Then there is Damn Small Linux, which fits on one of those credit card sized CDs. Not for your average Windows newbie, but wow, it just shows what you can do when you have such a good solid base to work off of.
In the end, Linux is a technically superior product that is only a little bit behind wrt usability - and the gap is closing very fast.
I am not sure that this still holds true. Perhaps it held true when the world was much more isolated, but I cannot think of any recent instances where the world suddenly iconized someone long since dead.
I think if Shakespeare was alive today we would all know about him and be enjoying his movies, and he would be very happy in his Hollywood mansion.
I mean, people used to have very hard-working days, but in their evenings, they probably had much more time to themselves than we do today. TV and computer games are the big culprits in today's world, stealing all of our free time and numbing our minds.
Its like another recent slashdot discussion about how abundance of something previously available in limited supply often causes unexpected results. We have an overabundance of entertainment, and we feast on it because we cannot help ourselves, thus sacrificing other more wholesome and productive pursuits.
I have not RTFA, but it makes sense to be that the spam response rates are high for the people that took the survey, simply because the *same people* that respond to meaningless surveys are the ones that respond to spam.
In other words, the survey technique has skewed the results, and is (like 99.81% of statistics), meaningless.