I can only assume, based on this latest news, that VIA is simply not interested in selling to market segments where "legacy" ports are still required. Fair enough. I'll stick to "real" motherboards, and VIA can stick to their goodies.
Well yeah, (or at least it's not a market for this release.) This is about like bitching that Alienware does not make $500 machines for college students to write their term papers on. If you want a low-power ITX, there is always the EPIA with serial and parallel.
the point about this board is its the -only- one of its size factor. if we were talking about ATX, you might have a point.
I guess I'm still not seeing how this complaint is valid or worthwhile. After all, the EPIA boards are 15cm X 15cm and have the full range of ports. I can't imagine an application where a whopping total of 1.2 inches is going to make all the difference.
I mean dang, why don't I start complaining that the EPIA lacks onboard sATA and RAID? They made a decision based on the market that they want to reach.
Chaos theory will only be called chaos theory until we expand our understanding to encompass the underlying phenomena that bridge the events that today seem unconnected.
Um, no.
Chaos theory has nothing to do with a lack of understanding of the underlying phenmonena. One of the first chaotic systems that was well studied after all was Isaac Newton's three body problem. Everyone knew how gravity worked between two bodies, but the greatest minds of two centuries could not figure out what happend with three or more bodies. Finally it was proven that except for a small handful of exceptions, there is no way to determine what three or more bodies will do to each other under the influence of gravity. (Fortunately for life on Earth, our solar system appoximates one of those handful of solutions.)
Chaos theory is not about the problems understanding the causes, but the problems predicting the effects.
The problem is not the writing, it's the editing. I think that the average published work goes through at least 6 revisions. As such, the ability to quickly find and edit text is a major bonus.
One of my more productive writing periods used vim as the editor of choice. I'd spend the morning banging out at least 1,000 words of pure shit. Eat lunch with a print-out and fountain pen, then rip through the edits in under a half hour, and spend the afternoon doing research. vim's ability to search through and change text is amazing. (The only reason why I don't continue to write in vim is that everyone wants word documents these days.)
How many times does someone need a regular expression search?
Nit 4: Sendmail. Sure. You could run sendmail, but why not look into qmail, written by djb. While you're there, check out djbdns if you need DNS services.
Actually, a bit further down they the author recommends postfix. But gee, there is just so much ground to cover here, splitting this up would be good.
I think that part of the problem is that there are large numbers of people who don't know they need it. I feel like a Cassandra for arguing that research data involving human participants should not be transmitted over the internet unencrypted, and should be stored encrypted in insecure environments (including the laptops that everyone hauls around everywhere.) It's not just credit card numbers, but just about any data that is covered by a confidentiality agreement.
However, I've never been able to convince anyone that the security is worth the trouble.
Well, yeah, they are assuming that you are using the electricity for more than just lighting. Solar lighting is easy. Running the rest of the office building is the hard part.
Re:The efficiency kills you, but KISS rules
on
Solar Window Panes
·
· Score: 1
Actually....
If you are going to use trick mirrors, it looks like the best way is to use a Stirling Engine to convert the solar energy to electricity. One example is Stirling Energy Systems but Discover just ran an article about the sunflower prototype from Energy Innovations that plans to break the magic $1/watt barrier. The sunflower uses reflective plastic petals to focus light onto a Stirling engine which generates electricity. This approach gets around some of the worst characteristics of PV cells such as sensitivity to the angle of light, high cost of material manufacturing, and upper limits on how much light can be converted.
I tend to make heavy use of glimpse if I need to find something. In my experience a good search and index tool works much better than spending time categorizing (and then trying to remember how I categorized that bit of information.)
Just spent about 6 minutes going through my spambox in search of false positives. The great thing about spam filtering is that the false positives stick out like a sore thumb in the midst of subject lines like "Can you satisfy her" and "&%IDU&*!". And 90% of the false positives are low-priority mail, like announcements for conferences I can't afoard to go to.
compared with modern multimedia communications Its a waste of bandwidth if all youre transmitting around the world is your own voice.
A problem with this is that computer users confuse bandwidth with throughput. In fact voice is a more conservative use of bandwidth than digital multimedia. This is the reason why telephone companies can offer universal voice access through the united states, but limited DSL access. (HAM digital modes such as CW, and PSK31 are even more conservative.)
Actually, one of the Melamine collectors handbooks, (believe it or not, there are people who run around trying to collect complete sets of brightly colored plastic dinnerware from the 50s) claims that the makers of Melmac had the Dod use Melamine dinnerwar. in above-ground nuclear weapons tests. Evidently the DoD was interested in the survivability of everyday materials and the producers of Melmac got free advertising on how indestrcuctable their product was.
Yeah, although I think there are problems with LaTeX as a format. Like the need to bundle multiple files together, and the fact that parsing a document depends on everyone having both the document, and the same version of the output stylesheet.
I really like the Open Office file format because it combines some of the features of structured document style in TeX but wraps up everything into a zip wrapper. I managed to build a simple sxw to text script in about 2 hours (about 90 minutes of which was spent tracking down an under-documented feature of the Python SAX API).
LaTeX is good, wonderful, even beautiful to the point where if I ruled the world, I would be using it exclusively. The big deal-killer with it is that more and more journals are wanting.doc. After spending a few weeks in search of a conversion filter from LaTeX + APA to RTF, I just jumped into OpenOffice.
One of the things that I think needs to be considered in comparing false-positive rates is not "can the false positive rate be eliminated" but "is the false positive rate lower than human sorting alone". As far as I know, nobody has really bothered to factor in human performance.
From the description it sounds like the $200 model uses a VIA mini-itx motherboard which includes integrated video and ethernet. I just puchased one to upgrade my FreeBSD box. The good news about these is that they are tiny (17cm X 17cm), have low power consumption and low heat output so they can be used anywhere. The bad news is limited expansion options (one ram slot+one pci slot), and a slow FPU. It's not going to win any performance awards but after all, I spent all of the last 3 years using a 350MHz Intel with few problems.
I'd love to use LaTeX but for one big problem. For the APA stylesheet there seems to be no output path to get to something someone else can use. With most of the major journals in my field requesting.doc submissions, this is not good.
It seems you haven't used the graphical SQL OOo query tool. It is exactly the same as in Access. OK I admit it is quite hidden in the interface.
On this, I must disagree. The OpenOffice database functionality is broken in many important ways. Can't make a form from a query that contains non-standard SQL, the database queries are rewritten in ways that are not necessarily helpful (or functional), subforms are only possible if you use the right driver, the JDBC driver sets all timestamp fields to 23:59. Frequent crashes.
I spent about 24 hours last week attempting to make an OO database form to view and add comments to the 120,000 database records that need to be coded for my dissertation. Finally I realized I would be much better off with a Python CGI script than OpenOffice. (Access got closer to what I need but I really don't have the time to learn VBA.)
Basically, Open Office database access is ok for basic applications and breaks quickly for more complex applications. I still use it exclusively but it as nowhere near being a replacement for Access.
I wouldn't use either Winzip or PKZip to send an encrypted zip file, because PGP is more universally known, and can give you 2048 bit encryption.
It should be noted that the 2048 bit encryption only applies to the public key which is a completely different beast. The body of the file is encrypted with CAST (at least in my version) that uses 40-128 bit keys.
But I've found that PGP is not well known at all (to the point where the last time I attempted to sign email by default, bunches of people complained about the PGPMime attachments.)
Not quite. One of the reasons why you compress before encryption is because good encryption algorithms produce output that can't be distinguished from random bytes. As a result, encrypted files can't be compressed (because there are no repeating fragments of any length.)
I don't know where you get the idea that the encrypting of files makes them much larger. For pretty much all encryption methods the output should just a little bit larger than the input. A file encrypted with AES 256 should be at most 255 bits longer than the original.
Most linux distributions seem to be leaning towards a complete desktop in a box approach. The BSDs lean more towards a minimal unix with everything else helpfully provided by packages and ports. (For example, bash is not installed by default, but adding it is trivial.)
Install
I found the default FreeBSD install to be a bit tricky. (Partly because I ran out of disk space and had to start again from scratch.) The FreeBSD install assumes that you know a bit about Unix and can read the instructions carefully. I'm told that Linux is an easier install.
Speed and power
YMMV. FreeBSD allegedly can take higher network loads. But, MySQL historically has not run as well under FreeBSD. (I've also ran into problems with threaded apache2.) Some anecdotal reports claim snappier desktop performance under FreeBSD.
Hardware support
Linux is ahead on new hardware. NetBSD runs on more platforms.
Community
Linux has a wider community. I've found support from FreeBSD groups to be pretty good.
My personal opinion is that I went with FreeBSD because of the better security record. With the exception of some minor glitches getting apache2 to run, I've been happy with it.
I think there is a concern here in that I've found that bayesian filtering works best because it is individualized to me. A shared database could be poisoned by a malicious user.
I've noticed this also. However, the nice thing about baysian filters is that they adapt along with the spammers. As spammers adopt new mispellings, the filter adapts to the new statistical model. Furthermore spammers can't do much about the features that result in a high ham score.
- The FreeBSD ports system is awesome. As far as I know, most Linux distributions don't have anything like it. I think one has something that is close but is based upon precompiled binaries. FreeBSD's ports you compile yourself, which takes longer but has benefits that I like. I think the various Linuxes are trying to "catch up" to FreeBSD in this regard but I could be wrong about the current state of affairs. Keeping apps (and the OS itself) up-to-date with tools like CVSup and portupgrade is sickeningly easy.
I agree. When I was a RedHat user, I found upgrading to be a bit spotty with rpms (easy to get into dependency hell.)
On the other hand, one of the other nice things about the ports system is that you don't have to compile from scratch all the time. For most of the trivial stuff I just use portupgrade -PP (upgrade only from packages) to save time.
Slashdot Mirror
I can only assume, based on this latest news, that VIA is simply not interested in selling to market segments where "legacy" ports are still required. Fair enough. I'll stick to "real" motherboards, and VIA can stick to their goodies.
Well yeah, (or at least it's not a market for this release.) This is about like bitching that Alienware does not make $500 machines for college students to write their term papers on. If you want a low-power ITX, there is always the EPIA with serial and parallel.
the point about this board is its the -only- one of its size factor. if we were talking about ATX, you might have a point.
I guess I'm still not seeing how this complaint is valid or worthwhile. After all, the EPIA boards are 15cm X 15cm and have the full range of ports. I can't imagine an application where a whopping total of 1.2 inches is going to make all the difference.
I mean dang, why don't I start complaining that the EPIA lacks onboard sATA and RAID? They made a decision based on the market that they want to reach.
Chaos theory will only be called chaos theory until we expand our understanding to encompass the underlying phenomena that bridge the events that today seem unconnected.
Um, no.
Chaos theory has nothing to do with a lack of understanding of the underlying phenmonena. One of the first chaotic systems that was well studied after all was Isaac Newton's three body problem. Everyone knew how gravity worked between two bodies, but the greatest minds of two centuries could not figure out what happend with three or more bodies. Finally it was proven that except for a small handful of exceptions, there is no way to determine what three or more bodies will do to each other under the influence of gravity. (Fortunately for life on Earth, our solar system appoximates one of those handful of solutions.)
Chaos theory is not about the problems understanding the causes, but the problems predicting the effects.
The problem is not the writing, it's the editing. I think that the average published work goes through at least 6 revisions. As such, the ability to quickly find and edit text is a major bonus.
One of my more productive writing periods used vim as the editor of choice. I'd spend the morning banging out at least 1,000 words of pure shit. Eat lunch with a print-out and fountain pen, then rip through the edits in under a half hour, and spend the afternoon doing research. vim's ability to search through and change text is amazing. (The only reason why I don't continue to write in vim is that everyone wants word documents these days.)
How many times does someone need a regular expression search?
Nit 4: Sendmail. Sure. You could run sendmail, but why not look into qmail, written by djb. While you're there, check out djbdns if you need DNS services.
Actually, a bit further down they the author recommends postfix. But gee, there is just so much ground to cover here, splitting this up would be good.
I think that part of the problem is that there are large numbers of people who don't know they need it. I feel like a Cassandra for arguing that research data involving human participants should not be transmitted over the internet unencrypted, and should be stored encrypted in insecure environments (including the laptops that everyone hauls around everywhere.) It's not just credit card numbers, but just about any data that is covered by a confidentiality agreement.
However, I've never been able to convince anyone that the security is worth the trouble.
Well, yeah, they are assuming that you are using the electricity for more than just lighting. Solar lighting is easy. Running the rest of the office building is the hard part.
Actually....
If you are going to use trick mirrors, it looks like the best way is to use a Stirling Engine to convert the solar energy to electricity. One example is Stirling Energy Systems but Discover just ran an article about the sunflower prototype from Energy Innovations that plans to break the magic $1/watt barrier. The sunflower uses reflective plastic petals to focus light onto a Stirling engine which generates electricity. This approach gets around some of the worst characteristics of PV cells such as sensitivity to the angle of light, high cost of material manufacturing, and upper limits on how much light can be converted.
I tend to make heavy use of glimpse if I need to find something. In my experience a good search and index tool works much better than spending time categorizing (and then trying to remember how I categorized that bit of information.)
Just spent about 6 minutes going through my spambox in search of false positives. The great thing about spam filtering is that the false positives stick out like a sore thumb in the midst of subject lines like "Can you satisfy her" and "&%IDU&*!". And 90% of the false positives are low-priority mail, like announcements for conferences I can't afoard to go to.
compared with modern multimedia communications Its a waste of bandwidth if all youre transmitting around the world is your own voice.
A problem with this is that computer users confuse bandwidth with throughput. In fact voice is a more conservative use of bandwidth than digital multimedia. This is the reason why telephone companies can offer universal voice access through the united states, but limited DSL access. (HAM digital modes such as CW, and PSK31 are even more conservative.)
Actually, one of the Melamine collectors handbooks, (believe it or not, there are people who run around trying to collect complete sets of brightly colored plastic dinnerware from the 50s) claims that the makers of Melmac had the Dod use Melamine dinnerwar. in above-ground nuclear weapons tests. Evidently the DoD was interested in the survivability of everyday materials and the producers of Melmac got free advertising on how indestrcuctable their product was.
Yeah, although I think there are problems with LaTeX as a format. Like the need to bundle multiple files together, and the fact that parsing a document depends on everyone having both the document, and the same version of the output stylesheet.
I really like the Open Office file format because it combines some of the features of structured document style in TeX but wraps up everything into a zip wrapper. I managed to build a simple sxw to text script in about 2 hours (about 90 minutes of which was spent tracking down an under-documented feature of the Python SAX API).
LaTeX is good, wonderful, even beautiful to the point where if I ruled the world, I would be using it exclusively. The big deal-killer with it is that more and more journals are wanting .doc. After spending a few weeks in search of a conversion filter from LaTeX + APA to RTF, I just jumped into OpenOffice.
One of the things that I think needs to be considered in comparing false-positive rates is not "can the false positive rate be eliminated" but "is the false positive rate lower than human sorting alone". As far as I know, nobody has really bothered to factor in human performance.
From the description it sounds like the $200 model uses a VIA mini-itx motherboard which includes integrated video and ethernet. I just puchased one to upgrade my FreeBSD box. The good news about these is that they are tiny (17cm X 17cm), have low power consumption and low heat output so they can be used anywhere. The bad news is limited expansion options (one ram slot+one pci slot), and a slow FPU. It's not going to win any performance awards but after all, I spent all of the last 3 years using a 350MHz Intel with few problems.
You can also correct this within open office (set the font for bullets to samething like Times New Roman)
I'd love to use LaTeX but for one big problem. For the APA stylesheet there seems to be no output path to get to something someone else can use. With most of the major journals in my field requesting .doc submissions, this is not good.
It seems you haven't used the graphical SQL OOo query tool. It is exactly the same as in Access.
OK I admit it is quite hidden in the interface.
On this, I must disagree. The OpenOffice database functionality is broken in many important ways. Can't make a form from a query that contains non-standard SQL, the database queries are rewritten in ways that are not necessarily helpful (or functional), subforms are only possible if you use the right driver, the JDBC driver sets all timestamp fields to 23:59. Frequent crashes.
I spent about 24 hours last week attempting to make an OO database form to view and add comments to the 120,000 database records that need to be coded for my dissertation. Finally I realized I would be much better off with a Python CGI script than OpenOffice. (Access got closer to what I need but I really don't have the time to learn VBA.)
Basically, Open Office database access is ok for basic applications and breaks quickly for more complex applications. I still use it exclusively but it as nowhere near being a replacement for Access.
I wouldn't use either Winzip or PKZip to send an encrypted zip file, because PGP is more universally known, and can give you 2048 bit encryption.
It should be noted that the 2048 bit encryption only applies to the public key which is a completely different beast. The body of the file is encrypted with CAST (at least in my version) that uses 40-128 bit keys.
But I've found that PGP is not well known at all (to the point where the last time I attempted to sign email by default, bunches of people complained about the PGPMime attachments.)
Not quite. One of the reasons why you compress before encryption is because good encryption algorithms produce output that can't be distinguished from random bytes. As a result, encrypted files can't be compressed (because there are no repeating fragments of any length.)
I don't know where you get the idea that the encrypting of files makes them much larger. For pretty much all encryption methods the output should just a little bit larger than the input. A file encrypted with AES 256 should be at most 255 bits longer than the original.
This is an good request. Briefly:
Package philosophy
Most linux distributions seem to be leaning towards a complete desktop in a box approach. The BSDs lean more towards a minimal unix with everything else helpfully provided by packages and ports. (For example, bash is not installed by default, but adding it is trivial.)
Install
I found the default FreeBSD install to be a bit tricky. (Partly because I ran out of disk space and had to start again from scratch.) The FreeBSD install assumes that you know a bit about Unix and can read the instructions carefully. I'm told that Linux is an easier install.
Speed and power
YMMV. FreeBSD allegedly can take higher network loads. But, MySQL historically has not run as well under FreeBSD. (I've also ran into problems with threaded apache2.) Some anecdotal reports claim snappier desktop performance under FreeBSD.
Hardware support
Linux is ahead on new hardware. NetBSD runs on more platforms.
Community
Linux has a wider community. I've found support from FreeBSD groups to be pretty good.
My personal opinion is that I went with FreeBSD because of the better security record. With the exception of some minor glitches getting apache2 to run, I've been happy with it.
I think there is a concern here in that I've found that bayesian filtering works best because it is individualized to me. A shared database could be poisoned by a malicious user.
I've noticed this also. However, the nice thing about baysian filters is that they adapt along with the spammers. As spammers adopt new mispellings, the filter adapts to the new statistical model. Furthermore spammers can't do much about the features that result in a high ham score.
- The FreeBSD ports system is awesome. As far as I know, most Linux distributions don't have anything like it. I think one has something that is close but is based upon precompiled binaries. FreeBSD's ports you compile yourself, which takes longer but has benefits that I like. I think the various Linuxes are trying to "catch up" to FreeBSD in this regard but I could be wrong about the current state of affairs. Keeping apps (and the OS itself) up-to-date with tools like CVSup and portupgrade is sickeningly easy.
I agree. When I was a RedHat user, I found upgrading to be a bit spotty with rpms (easy to get into dependency hell.)
On the other hand, one of the other nice things about the ports system is that you don't have to compile from scratch all the time. For most of the trivial stuff I just use portupgrade -PP (upgrade only from packages) to save time.