My dad gets AOL for $3.95 a month; apparently his employer decided to sponsor employee access to the Internet to save on the cost of distributing paper copies of stuff like the benefit-plan description. It's not clear whether the employer pays anything to AOL.
because they never got it to compile (it was for another architecture, right?) and decided to just delete it instead. This brings up a question: does a software developer have a legal duty to keep copies of all revisions and attempts he makes with his code, even those that don't work or turn out to be useless?
Well, sure, a lot of programs run just fine on newer versions of Windows, and if a program does not keep running, it may be that it was not designed right in the first place.
Speaking of backwards compatibility, I have a story that tops yours. Back in 1988 or 1989, my dad would type his research-papers with a proprietary system called Mass-11 that ran under plain DOS (his PC was an 8088 clone), export them to ASCII files, and run them through TeX. Just a month or so ago, we tested the programs on his Windows ME box, and it seemed to do everything he might want; however, we didn't try printing to the printer he used to use, because he also has MikTeX (which can print to his DeskJet or PDF) installed on that system. I've also run the program on an NT 4.0 box, so I doubt it would have any problems on XP Pro. That's a 15-year program useful lifetime. Of course, my dad would probably be better off just making sure he's exported everything he might possible want to copy and then junk the program.
If a program stuck to the standard C library or the documented DOS API, it probably kept working from then 'till now; same thing for Windows program that stuck to the core API. However, in the DOS/Windows/Visual Basic/.NET programming milieu there's long been an attitude that one needs to use obscure or undocumented APIs to produce good programs. Sure, some of this came from sources outside of Microsoft, but even today articles on MSDN tend to encourage writing to the latest-and-greatest version of Windows using non-backwards-compatible toolkits or still-developing platforms. (This one, for instance, talks about "Avoiding the Win32 API"!). Now, oficially.NET is an ECMA standard, just like
JavaScript, but even MSDN encourages Windows-centric ways of doing things (which is bad because of this question about Registry support, for instance). I guess the documentation for gets() has an OK warning to use fgets() instead, but I think the "BUGS" section of the corresponding UNIX manpage
is better.
The PowerPC chip was designed with features to make it easy to port or emulate x86 code, like a memory-access system that could be either big- or little-endian. Even so, NT workstations based on it were never a consumer-market item, and probably were never widely used. (Actual experience, anyone?)
Windows at present is mostly based on the 32-bit Intel architecture. Microsoft did its worst dirty tricks in the last dying days of the segmented 16-bit architecture, using DOS dominance to get market share for its 32-bit attempt. It's going to have to chose between AMD-64 and Intel-64 anyway, or support both, and binary application developers will need to make the same choice, so I guess the submitter would argue that PPC-64 (which has been around longer) is a viable option. However, there's a big movement away from software that's tied down to one platform or another, which is good for Linux, Java, and all the other OS, hardware and software vendors, programmers, and users.
The limited adoption and big troubles implementing Wine suggests to me that there would be little interest in a Microsoft port of Windows to yet another architecture. Windows 95 was probably the most-memorable MS-Windows version ever, and yet Microsoft has had to fragment even that identity to keep up its sales, starting with that crazy desktop in XP. The claim that Windows has excellent backward compatibility is bogus, too; for instance, the copy of TeraTerm that I carry around on a floppy has never worked on any NT2k or later system I've touched, and the default installation of Microsoft Word can't read files created by any version of Microsoft Works. I could contiue this rant...
If the thief has physical access to the machine, nothing short of encryption is going to prevent him or her from getting at your data.
If you had read the article, you would have seen that this device can protect the BIOS password, Windows passwords and a hard-drive encryption suite. It sounds like a substantial obstacle to the typical thief. Of course, you should probably still be careful with really-sensitive data; PGP and daily virus scanning would be a good idea...
According to the article, Sun says its tecnology is "patent pending"; to me, it sounds like they want to keep tight control of it. Of course, patents do expire eventually...
Your method is still subject to traffic analysis, viz.: A is a known terrorist, and A has sent email/called/chatted via AIM/met B and C; therefore we can wiretap all communications between B and C. However,
Party A could send out thousands of "customized" advertisements for an obviously fraudulent product by e-mail, each with hidden, seemingly random string of text at the end. Most recipients would just hit the "delete" key, but B and C would run the seemingly-random text through some algorithm with some key and recover the plaintext of the message. Party A would use standard cloaking techniques just as well as any other spammer, and parties B and C would just be checking their e-mail and "only finding spam".
Who says spammers aren't actually terrorists in disguise?
As for java, it's one of the primary languages for easy interface with databases
perl is great for databases, too; look into the DBI:: classes some time. The thing that makes Java useful is the commercial toolkits that easily build GUI forms, JavaBeans, etc.; however, the main interface that big companies need to their database is either their website or a simple locked-down interface to a transactional backend. PHP and ASP easily compete with JSP for the first, and a perl program running in an SSH instance might be the best environment for new development of the other. (However, I don't have a production system to back that claim up.)
This would probably be a good exercise for an introductory network-programming class: write an NFS or SMB server in java or perl. It would be a dog for performance, though, and I'm not sure how to store/retrieve files bigger than max_packet_size. The other disadvantage is that this "filesystem" would also work on Windows.
HTTPS is good for transport-level encryption, but it's still vulnerable to man-in-the-middle attacks or a compromised certificate authority. I'm surprised he's not using some form of PGP; he could presumably get a free certificate from CAcert, but there are only three members of that organization in New Zealand who can assure his identity.
The security of the protocol itself is just fine; it's the surrounding factors that make it hard to use, especially for someone who hasn't thought about these things well enough ahead of time.
I tried all the links on the test page, but I just got a message-box saying "shell is not a registered protocol". Could it be that I'm using Mozilla 1.4 on Linux?
Corporations are required to submit financial statements to the SEC and tax returns to the IRS. If there are inconsistencies, shareholders, creditors or the government can ask in court to see the accounting records, and even internal correspondence (e.g., Enron). This is no more than a similar documentation requirement.
perdoc is, in my experience, the best reference for an API. I would never say it's worse than any part of MSDN. Then again, the Win32 stuff on MSDN is often an edited version of what was there for Windows 3.0 (1990 or before), while the manpages for core commands and functions on UNIX are typically repackaged or rewritten versions of documentation from the AT&T codebase (1982? 1975?). Some documentation is very similar, or even the same, in manpages and MSDN; for instance, most of the OpenGL functions, when looked up in MSDN or the manpages on Solaris, are described word-for-word the same.
I keep around a copy of Borland C++ 5.0 partly just because of the.HLP files documenting various APIs. When trying to learn about an unfamiliar API, a website (with waits between pages) just isn't as efficient as a fat-client-ish help browser. Windows help was good; 'man'+'less' is excellent, except that it doesn't store any sort of hierarchy. However, a tree-view can actually make it harder to grok a system; sooner or later I have to read something more-or-less from beginning to end to understand it.
Every time I try to go to a SourceForge.net site, I get redirected to a 404 page on "portland.co.uk", and their page even shows up in the ad-boxes on Slashdot; is there any connection here? Or has someone been messing with the DNS locally (at Wayne State University, Detroit, MI, USA)?
The/etc structure should be emulated and config info should be left to flat file structures.
I like Linux much better than Windows, but I think the Registry, or something like it, is one thing that would be useful. Right now there are probably 20 different file-formats running around under/etc and in ~/.whatever, each one requiring a slightly different algorithm to parse. Furthermore, depending on the exact options when something was compiled, it might actually be looking in/usr/local/etc or/usr/share/whatever or...
With the Registry, about 5 system-calls are all that's needed to read/test existence/write data. Sure, "regedit" doesn't expose all the cross-references between different keys, but you don't have to worry that you deleted a tab by mistake when changing a value somewhere, and you can drill down to one value pretty quickly; if you know what key you're looking for, you don't need to do grep 'keyword' `find/etc -type f` or anything like that.
Another problem with Linux configuration-files is that they tend to involve multi-tiered directories of little, oddly named files. I still haven't figured out how to copy an icon from one user's GNOME menus to another's; under Windows NT, I would just do drag-and-drop between the corresponding Start Menu folders. However, CDE is the worst: applications have to be represented by a zero-length executable file in one directory AND a corresponding type description in another!
I'm not saying that current configuration-files should be scrapped, just that (A) adding to new formats should be frowned on, (B) programs should reuse other programs' configuration-files as much as possible, and (C) someone should write a regedit-like tool for Linux that parses and even edits a lot of the common files. It should show data in a format something like
Root
System
passwd
user1
id=123
name=User 1
hosts
127.0.0.1=localhost
192.168.1.10=foo.mydomain.org
Users
dave
john
CurrentUser
passwd (one entry, only chfn-able entries writable)
ssh
identity
authorized_keys
known_hosts
config
wine
CurrentUser
config
mime
menus
forward
odbc
, etc., etc. The program should pop up a text editor for multi-line text in a file it doesn't recognize, and there should be simple perl calls to read and set the configuration data, too.
It's really ridiculous. Employers ask you to sign employee agreements which are utterly ridiculous in their scope and demands but it's either sign the line or take your chances at McDonald's.
My present job is a good example, although it's non-IT-related. When I got hired, I thought it was for a full-time position. At the first staff meeting, we were all told to work a 7.5-hour day, and "no overtime". A couple months later, that got cut to a 6.5-hour day. At least I still get paid more per hour than at McDonalds...
I found out that the same organization was hiring for their IT department, so I sent in my resume, but I haven't been interviewed yet. My boss mentioned once that her husband is an unemployed IT professional, too.
A few months ago I went through the Army recruitment
process and expressed an interest in some MOSes that
would have required top-secret clearance. During the "interview", some things came up that made the interviewer think I couldn't get clearance; then she got angry and misinterpreted my medical file to completely disqualify me from military service. She told me that a full top-secret background check costs over $60 thousand, but I'm not sure whether I believe her.
Does anyone know whether the questions on the Army security interview are supposed to be secret? If not, I might want file a FOIA to get them and refresh my memory about exactly what I "said". I was looking at the FBI security-clearance application more recently, and it was different in a lot of ways.
Another one of their schemes offers to download a program that will automatically sync your system time, which is useful to most people who notice that consumer PCs are usually pathetic at keeping a system time.
The Linux fileserver/firewall on my network keeps its time synchronized with ntpdate. All the Windows boxes run a batch script that does NET TIME \\[linux box]/SET/YES whenever anyone logs in. The setup seems to work fine; I also see that ther is a version of NTP for winNT.
Regardless, I say if you registered the domain it belongs to you. What about that guy who was able to pick up hotmail.co.uk because Microsoft forgot to renew it? I say he should be under no obligation to return it.
Unless you don't have any good rights in the domain name in the first place...
I live in Macomb county, Michigan, USA. The county sheriff's department had a website at macombsheriff.com, but the volunteer maintainer got tired of his duties, closed down the website, and sold the doamin to a third-party domain dealer. I'm thinking about writing to the county commissioners and telling them they should have used macomb.co.mi.us/sheriff/ as their main website instead.
(While I'm ranting about this, what's with all the junk in the.US domain? The other day I got about 15 pieces of spam all advertising two websites, both in.US, run by the same guy in the Netherlands and selling mail-order viagra or something. Right now, it's almost as bad as.biz. I guess all the government entities grabbing.com names have pushed the shady enties right on out.)
Consider the number 10. In order to read that as "ten", you have to read the 0 first. That's on the right. The 0 is in the ones place. Then you have to read the 1, which is in the tens place. Then you have to put this information together and get "ten", but you have to read it left-to-right to be able to read "ten". So, numbers are read in both directions.
It's possible to find the value of a number reading one digit at a time from right-to-left or from left-to-right. Only when saying it out loud is it necessary to read the whole number before starting to say it. In fact, I don't actually read each digit twice when reading a large number; I just count groups of 3 and then start at the left. However, I suppose that this means it would be impossible to design a finite state machine to to translate sequences of decimal numbers into their spoken English forms or vice-versa; the second language is computable but not regular. Is it recursive?
Incidentally, Arabic writes numbers highest-value-to-lowest, left-to-right even though text is written right-to-left. Farsi probably does the same thing.
For some reason Yahoo doesn't make links to sites they mention in their stories, so I'll fix that part for them. The second paragraph should read:
A list published in the Federal Register includes newkach.org, kahane.org, kahane.net, kahanetzadak.com as aliases for the Jewish group Kahane Chai or Kach, which is suspected of organizing attacks on Palestinians...
If it's illegal to give money to these sites, is slashdotting them a patriotic act?
It's amazing how many super cool random people are running around suggesting using OpenNIC, which, of course, won't do a D--- F--- THING. Anyone who suggests an alternate root has demonstrated they have no knowledge of how DNS works at the topmost level
Not in and of itself, but we hope OpenNIC will quickly move to sanitize the domain-information they pass on, which should be easy with the patches that have been released. Better to fix the problem at one point than at thousands.
Even before Verisign did this, all the recent browsers had a 'search on mispelling' feature on the address bar -- IE (MSN), Mozilla (Google), Netscape 6 (Netscape.com). VeriSign just wanted to jump into the market, and now other ISPs will follow suit.
Slashdot Mirror
My dad gets AOL for $3.95 a month; apparently his employer decided to sponsor employee access to the Internet to save on the cost of distributing paper copies of stuff like the benefit-plan description. It's not clear whether the employer pays anything to AOL.
because they never got it to compile (it was for another architecture, right?) and decided to just delete it instead. This brings up a question: does a software developer have a legal duty to keep copies of all revisions and attempts he makes with his code, even those that don't work or turn out to be useless?
Speaking of backwards compatibility, I have a story that tops yours. Back in 1988 or 1989, my dad would type his research-papers with a proprietary system called Mass-11 that ran under plain DOS (his PC was an 8088 clone), export them to ASCII files, and run them through TeX. Just a month or so ago, we tested the programs on his Windows ME box, and it seemed to do everything he might want; however, we didn't try printing to the printer he used to use, because he also has MikTeX (which can print to his DeskJet or PDF) installed on that system. I've also run the program on an NT 4.0 box, so I doubt it would have any problems on XP Pro. That's a 15-year program useful lifetime. Of course, my dad would probably be better off just making sure he's exported everything he might possible want to copy and then junk the program.
If a program stuck to the standard C library or the documented DOS API, it probably kept working from then 'till now; same thing for Windows program that stuck to the core API. However, in the DOS/Windows/Visual Basic/.NET programming milieu there's long been an attitude that one needs to use obscure or undocumented APIs to produce good programs. Sure, some of this came from sources outside of Microsoft, but even today articles on MSDN tend to encourage writing to the latest-and-greatest version of Windows using non-backwards-compatible toolkits or still-developing platforms. (This one, for instance, talks about "Avoiding the Win32 API"!). Now, oficially .NET is an ECMA standard, just like
JavaScript, but even MSDN encourages Windows-centric ways of doing things (which is bad because of this question about Registry support, for instance). I guess the documentation for gets() has an OK warning to use fgets() instead, but I think the "BUGS" section of the corresponding UNIX manpage
is better.
Windows at present is mostly based on the 32-bit Intel architecture. Microsoft did its worst dirty tricks in the last dying days of the segmented 16-bit architecture, using DOS dominance to get market share for its 32-bit attempt. It's going to have to chose between AMD-64 and Intel-64 anyway, or support both, and binary application developers will need to make the same choice, so I guess the submitter would argue that PPC-64 (which has been around longer) is a viable option. However, there's a big movement away from software that's tied down to one platform or another, which is good for Linux, Java, and all the other OS, hardware and software vendors, programmers, and users.
The limited adoption and big troubles implementing Wine suggests to me that there would be little interest in a Microsoft port of Windows to yet another architecture. Windows 95 was probably the most-memorable MS-Windows version ever, and yet Microsoft has had to fragment even that identity to keep up its sales, starting with that crazy desktop in XP. The claim that Windows has excellent backward compatibility is bogus, too; for instance, the copy of TeraTerm that I carry around on a floppy has never worked on any NT2k or later system I've touched, and the default installation of Microsoft Word can't read files created by any version of Microsoft Works. I could contiue this rant...
If you had read the article, you would have seen that this device can protect the BIOS password, Windows passwords and a hard-drive encryption suite. It sounds like a substantial obstacle to the typical thief. Of course, you should probably still be careful with really-sensitive data; PGP and daily virus scanning would be a good idea...
According to the article, Sun says its tecnology is "patent pending"; to me, it sounds like they want to keep tight control of it. Of course, patents do expire eventually...
Party A could send out thousands of "customized" advertisements for an obviously fraudulent product by e-mail, each with hidden, seemingly random string of text at the end. Most recipients would just hit the "delete" key, but B and C would run the seemingly-random text through some algorithm with some key and recover the plaintext of the message. Party A would use standard cloaking techniques just as well as any other spammer, and parties B and C would just be checking their e-mail and "only finding spam".
Who says spammers aren't actually terrorists in disguise?
perl is great for databases, too; look into the DBI:: classes some time. The thing that makes Java useful is the commercial toolkits that easily build GUI forms, JavaBeans, etc.; however, the main interface that big companies need to their database is either their website or a simple locked-down interface to a transactional backend. PHP and ASP easily compete with JSP for the first, and a perl program running in an SSH instance might be the best environment for new development of the other. (However, I don't have a production system to back that claim up.)
This would probably be a good exercise for an introductory network-programming class: write an NFS or SMB server in java or perl. It would be a dog for performance, though, and I'm not sure how to store/retrieve files bigger than max_packet_size. The other disadvantage is that this "filesystem" would also work on Windows.
The security of the protocol itself is just fine; it's the surrounding factors that make it hard to use, especially for someone who hasn't thought about these things well enough ahead of time.
I tried all the links on the test page, but I just got a message-box saying "shell is not a registered protocol". Could it be that I'm using Mozilla 1.4 on Linux?
As for me personally, I never have watched the Superbowl and probably never will, even though I'm an 18-to-30-year-old male. Once in a while I watch spanish-language TV or anime with my wife. On the radio, I tend to switch between WNZK (Arabic- and Spanish-language programming), the local NPR station, and CBC radio 2/A> (classical music broadcast from Canada). Obviously this requirement won't affect Canada, and I thought NPR already made a permanent archive of their broadcasts available.
I keep around a copy of Borland C++ 5.0 partly just because of the .HLP files documenting various APIs. When trying to learn about an unfamiliar API, a website (with waits between pages) just isn't as efficient as a fat-client-ish help browser. Windows help was good; 'man'+'less' is excellent, except that it doesn't store any sort of hierarchy. However, a tree-view can actually make it harder to grok a system; sooner or later I have to read something more-or-less from beginning to end to understand it.
Every time I try to go to a SourceForge.net site, I get redirected to a 404 page on "portland.co.uk", and their page even shows up in the ad-boxes on Slashdot; is there any connection here? Or has someone been messing with the DNS locally (at Wayne State University, Detroit, MI, USA)?
I like Linux much better than Windows, but I think the Registry, or something like it, is one thing that would be useful. Right now there are probably 20 different file-formats running around under /etc and in ~/.whatever, each one requiring a slightly different algorithm to parse. Furthermore, depending on the exact options when something was compiled, it might actually be looking in /usr/local/etc or /usr/share/whatever or ...
With the Registry, about 5 system-calls are all that's needed to read/test existence/write data. Sure, "regedit" doesn't expose all the cross-references between different keys, but you don't have to worry that you deleted a tab by mistake when changing a value somewhere, and you can drill down to one value pretty quickly; if you know what key you're looking for, you don't need to do grep 'keyword' `find /etc -type f` or anything like that.
Another problem with Linux configuration-files is that they tend to involve multi-tiered directories of little, oddly named files. I still haven't figured out how to copy an icon from one user's GNOME menus to another's; under Windows NT, I would just do drag-and-drop between the corresponding Start Menu folders. However, CDE is the worst: applications have to be represented by a zero-length executable file in one directory AND a corresponding type description in another!
I'm not saying that current configuration-files should be scrapped, just that (A) adding to new formats should be frowned on, (B) programs should reuse other programs' configuration-files as much as possible, and (C) someone should write a regedit-like tool for Linux that parses and even edits a lot of the common files. It should show data in a format something like
user1
id=123
name=User 1
Users
CurrentUser
wine
mime
menus
forward
odbc , etc., etc. The program should pop up a text editor for multi-line text in a file it doesn't recognize, and there should be simple perl calls to read and set the configuration data, too.
My present job is a good example, although it's non-IT-related. When I got hired, I thought it was for a full-time position. At the first staff meeting, we were all told to work a 7.5-hour day, and "no overtime". A couple months later, that got cut to a 6.5-hour day. At least I still get paid more per hour than at McDonalds...
I found out that the same organization was hiring for their IT department, so I sent in my resume, but I haven't been interviewed yet. My boss mentioned once that her husband is an unemployed IT professional, too.
Does anyone know whether the questions on the Army security interview are supposed to be secret? If not, I might want file a FOIA to get them and refresh my memory about exactly what I "said". I was looking at the FBI security-clearance application more recently, and it was different in a lot of ways.
The Linux fileserver/firewall on my network keeps its time synchronized with ntpdate. All the Windows boxes run a batch script that does NET TIME \\[linux box]
Agreed... this website isn't just funelling traffic, it's actually spamming the domain-name hierarchy.
Unless you don't have any good rights in the domain name in the first place...
I live in Macomb county, Michigan, USA. The county sheriff's department had a website at macombsheriff.com, but the volunteer maintainer got tired of his duties, closed down the website, and sold the doamin to a third-party domain dealer. I'm thinking about writing to the county commissioners and telling them they should have used macomb.co.mi.us/sheriff/ as their main website instead.
(While I'm ranting about this, what's with all the junk in the .US domain? The other day I got about 15 pieces of spam all advertising two websites, both in .US, run by the same guy in the Netherlands and selling mail-order viagra or something. Right now, it's almost as bad as .biz. I guess all the government entities grabbing .com names have pushed the shady enties right on out.)
It's possible to find the value of a number reading one digit at a time from right-to-left or from left-to-right. Only when saying it out loud is it necessary to read the whole number before starting to say it. In fact, I don't actually read each digit twice when reading a large number; I just count groups of 3 and then start at the left. However, I suppose that this means it would be impossible to design a finite state machine to to translate sequences of decimal numbers into their spoken English forms or vice-versa; the second language is computable but not regular. Is it recursive?
Incidentally, Arabic writes numbers highest-value-to-lowest, left-to-right even though text is written right-to-left. Farsi probably does the same thing.
A list published in the Federal Register includes newkach.org, kahane.org, kahane.net, kahanetzadak.com as aliases for the Jewish group Kahane Chai or Kach, which is suspected of organizing attacks on Palestinians...
If it's illegal to give money to these sites, is slashdotting them a patriotic act?
Not in and of itself, but we hope OpenNIC will quickly move to sanitize the domain-information they pass on, which should be easy with the patches that have been released. Better to fix the problem at one point than at thousands.
Even before Verisign did this, all the recent browsers had a 'search on mispelling' feature on the address bar -- IE (MSN), Mozilla (Google), Netscape 6 (Netscape.com). VeriSign just wanted to jump into the market, and now other ISPs will follow suit.
That looks like the list of nicks at some 'leet board... No wonder this is what gets traded on KaZaa.