Wow... I could have written what you wrote, that hits really damned close to home. Two differences:
* I find that occasionally my dreams do produce practical solutions to programming problems, occasionally forcing me to just get up and try it out - because otherwise I'll toss and turn and obsess over it.
* I no longer get into the zone much anymore, as I have two toddlers that dislodge me very regularly (for better or for worse... mostly for better on the human side).
Yep. Librarians and archivists aren't stupid -- that's why we have invented digital replication systems like Lots of Copies Keep Stuff Safe. From the site:
Libraries are using the LOCKSS Program to build libraries! With publishers, our community is working to retain libraries as long-term memory organizations in the electronic environment.
People with responsibility for scholarly assets agree: digital preservation is important. With your help, librarians and publishers are asking two fundamental questions: From this moment on, who will have custody of societies' electronic information? From this moment on, who will control and govern societies' electronic archival assets?
Join us! The community is working to insure important scholarly assets remain available in a distributed, self-repairing, robust, digital preservation system.
LOCKSS is OAIS compliant, LOCKSS migrates content forward in time, and LOCKSS continually audits and repairs the content. LOCKSS is open source software -- the system is freely available for you to examine and use.
We understand the importance of preserving cultural memory. LOCKSS is one way that we can cooperatively protect digital collections from physical calamity, abandoned formats, economic hardship, changing political climates...
Yes, really. I think its fair to say that this particular set of libraries (almost the entire set of libraries in the state of Georgia) has staked its future on an open-source library system: they're staking their future not just on the code they've developed and released to open source, but also on Linux, Apache, PostgreSQL, Perl, SpiderMonkey, Mozilla/XUL, Jabber, lib-dbi / lib-dbd, and umpteen Perl modules. Take away the OSS pieces that make up that system, and you take away their ability to function as a library. So yes, their future within the state of Georgia depends on OSS.
Turnkey systems may have all kinds of problems and be expensive, but the problems can be solved with money.
Of course, the nice thing with an open source ILS is that there is one problem that cannot be solved with money: if your vendor happens to go out of business, or decides to stop supporting your product, or decides to hike your fees immensely and you don't have a good means of exporting all of your data (including serials and acquisitions) to another system, then you're basically screwed. And given all of the consolidation and private equity investments going on in the library system vendor world at the moment, I'm not particularly confident about the viability of any of the current vendors. Open source at least gives you the assurance that you are not locked in with a given product; you own your data, and you can extend or enhance the system if you have the skills in-house or are willing to fund the skills.
And what makes more sense: having one hundred libraries paying $10K - $25K or more per year in basic support fees to vendors, or investing those same funds (or equivalent resources) towards a common library system that can be shared freely? Acquisitions and serials are tricky, sure, but you can make a lot of headway with that kind of common resource available. Hey, look: one institution has already joined GPLS to help develop an acquisitions module for Evergreen. And yes, the open source solution requires support resources: but proprietary solutions require support personnel as well... who, at least in the case of our library system, are each required to take a $3K training course to be certified as a system administrator, and have to take another $3K training course if they want to touch the API that the vendor has made available for the system.
I'm a systems librarian, so I claim to know of what I speak.
Most of it shouldn't even need to be converted. It should be in MARC Bibliographic format, which is generally fairly easy to transfer between databases.
This is true, as far as the bibliographic information goes. There are lots of open-source packages for working with MARC records, like pymarc (Python) or File_MARC (PHP). But the rest of the system is proprietary: holdings records, (which copies do you hold, in which locations, and where is that copy currently - loaned out, lost, on reserve, etc), circulation records, user records, acquisitions records. Sure, it's all just a database schema mapping exercise, if your vendor's license allows you to touch that data directly. Sadly, the past generation of libraries seems to have accepted vendor lock-in as a matter of course; a mistake that we're paying for now and which led directly to the development of Evergreen.
But really, let's be realistic. The major OPAC package is Voyager, which runs on top of Oracle, so runs on anything that runs Oracle. Libraries that don't have Voyager are pretty much all just wishing they could afford it (and the Oracle licenses).
Wow. This is just so wrong that I don't know where to begin.
First, Voyager is far from the market leader (in either usable interfaces or in market share). See
Second, the underlying database doesn't mean a thing if you aren't given the APIs to actually modify or extend your primary application, unless you're willing to reimplement the entire application -- in which case, why bother paying for a library system in the first place. And in most cases, when the vendor has made an API available, you have to pay extra fee per potential developer to receive the documentation and to be eligible for paid support for their API (which, of course, is an additional support fee over and above your standard support fees).
Third, most librarians I know couldn't care less about what technology their system is built on. They're focused on providing the best possible service to their users. Over the past few years, the library community has started to realize that there are some pretty cool Web interfaces out there in the wild that their vendors aren't providing for us. So we've been going through exercises like NCSU's use of Endeca (on the proprietary side) and Koha, Evergreen, and WPopac (on the open-source side) to try and correct the situation.
Librarians rock, you know.
Sure. Check out OSS4LIB for a list of different open-source library systems. On the smaller end of the scale, OpenBiblio, PhpMyLibrary, and Emilda get mentioned a fair bit.
It's a lot more complex than just running rpm against a tarball. This application depends on Apache, PostgreSQL, a Jabber server, libmemcached, CVS versions of SpiderMonkey and the lib-dbi / lib-dbd packages, and a host of Perl modules. Georgia's installation runs on top of 25 servers. Getting a secure, stable system up and running was understandably the Evergreen team's first priority; making the source and as much documentation available as they already have was a courtesy that they didn't even have to extend. However, they've gone further than that: they have made a commitment to open source, and librarians-who-do-development / developers-for-libraries thank them for that!
The project is still in its infancy as far as growing outside of Georgia. That's why a few of us are starting by trying to build Open-ILS (aka Evergreen) on our own, following the existing installation instructions. We've already been able to tease out a few more details in the process of trying to create distribution-specific instructions... not that the developers are trying to keep installation hard, it's just that they've lived and breathed this for the last year and a half and so some details are second-nature to them.
I have been quite impressed with the Evergreen team's technical capabilities and commitment to open source. They have contributed patches to the upstream libdbi and libdbd packages as a matter of course, and they pull in a ton of Perl modules that are either not packaged in most distributions, or on which they have dependencies on newer versions than are available in current distros. The team is committed to a transparent open source model comparable to the Linux or PostgreSQL development models and have posted a first draft of that model for comment.
So, the first hurdle is to set up an Evergreen environment outside of Georgia. I've come very close with Ubuntu and Gentoo, but have one more hurdle to clear. All the way along, the developers have been extremely responsive to my questions.
Once we nail down the "gotchas", my personal goal is to create a VMWare image that can be easily redistributed for demonstration purposes. I agree that packaging is obviously going to be an important aspect of the project, but it naturally has to follow a well-documented manual install process.
Evergreen is going to spread outside of Georgia, and it's going to spread fast. The University of Windsor has already announced their intention to work with Evergreen to build an acquisitions system.
Hmm. Now I come back, view the same long deeply nested thread, and it's all neat and pretty, just like the deities intended.
Either the Slashdot devs were working hard on correcting problems on their side, or IE7 is a little bit shaky. And we all know that Cmdr Taco et al are members of the Cult of Slack, and that Microsoft is pure evil, so...
Poor Slashdot. You guys go to all of the trouble of switching to clean tags + CSS, and Microsoft goes and updates their browser so that nested levels of comments render completely unintelligibly with text splashed laughably about.
I suspect Microsoft fears you, Slashdot hordes. Although you would have thought that the IE7 developers might have tested how Slashdot renders before releasing an update that is sure to incite fear and loathing in the masses gathered here (just like witches at black masses -- oh lord yeah!).
DB2 really isn't too hard to install on Ubuntu -- see the DB2 HOWTO for instructions.
One good reason to use DB2 is that neither MySQL nor PostgreSQL can scale to the same size of database as DB2. Using DB2's "Distributed Partitioning Facility", or DPF, with DB2 Enterprise Server Edition you can partition data over hundreds of machines using a shared-nothing approach. This means you can create a data warehouse of a hundred terabytes that 1) appears to your applications as just a single database 2) retrieves query results blazing fast (this is the "shared nothing" part; each machine knows what data it is responsible for, so a query that comes in to the database is automatically distributed to the responsble machines, which go off and do their work, then return the results to a single co-ordinator that pipes them back to the application -- it's like RAID-0 for databases).
It's certainly possible to create a multi-terabyte database using MySQL or PostgreSQL -- but you're going to be waiting way the hell longer to get a result back from your queries.
Reason 2: You get support from IBM. IBM lives and dies by its enterprise level support -- 24x7 customer service around the world. Sure, you can purchase support from MySQL AB or from some PostgreSQL shop, but it's not going to approach the level of service that you'll get from IBM. In fact, to get a level of support comparable to what IBM offers for DB2 Express at $4874 for the first year and $1218/year for support after the first year, you would need to buy MySQL Network Gold for $2995/year. So after two years, going with DB2 Express will actually cost you less than half of what MySQL support will cost!
... and you should be done before breakfast, let alone lunch.
There's a DB2 HOWTO at the Linux Documentation Project that provides detailed instructions for many popular Linux distributions, including special prereqs or workarounds for distros like Gentoo.
You might ask "why?" -- and the answer is the same as if you asked "Why are companies still selling Fortran compilers?": a whackload of existing, working code and a build environment that would cost mega-bucks to port to a new set of tools.
It would be interesting to see how VisualAge compares on ChaoticCoyote's benchmark. I hope IBM contacts him to get into the next edition of his benchmarks.
Declaring something "useless" means nothing if you haven't backed up your opinon with some rationales. You don't like/media because Redhat put cdroms in/mnt/cdrom? Fine, say so. Or you don't like it because it has too many vowels and should be/mda? Okay, great. But back up your opinion with some rationale. Otherwise it's just an assertion that takes up space.
Oh, and provide your rationale to those (like Rusty and Dan) who actually set the standards. Whining about it on Slashdot is hardly the way to achieve any change.
Here's what the FHS says about/media, by the way:
/media : Mount point for removeable media
Purpose
This directory contains subdirectories which are used as mount points for removeable media such as floppy disks, cdroms and zip disks.
Rationale
Historically there have been a number of other different places used to mount removeable media such as/cdrom,/mnt or/mnt/cdrom. Placing the mount points for all removeable media directly in the root directory would potentially result in a large number of extra directories in/. Although the use of subdirectories in/mnt as a mount point has recently been common, it conflicts with a much older tradition of using/mnt directly as a temporary mount point.
/opt comes from the Filesystem Hierarchy Standard (FHS). A lot of what the LSB tries to do is codify existing practice or point to existing standards wherever possible -- doesn't make sense to create something entirely from scratch if nobody is going to adopt it. The FHS is just one example of where the LSB pointed to an existing standard and said "This be good!"
Given that the FHS is headed up by notables Rusty Russell and Dan Quinlan, I've got a lot of confidence in their judgement.
When even Microsoft is admitting that Linux server shipments are growing faster then Microsoft server shipments, that's news.
Can anyone track down the original Gartner report that indicated 50% of server sales would be Linux by 2008? The linked article just mentions the Gartner report (and all-important statistic) in passing, but doesn't provide a proper reference for fact-checkers. Google didn't do the trick for me, it did turn up an article about an IDC report released in June 2004 that predicted Linux server shipments would rise to 29% in 2004, a fairly significant difference.
Novell's stock
is looking pretty attractive at $5.80, given that they're trading close to their 52-week low and now own SuSE, one of Red Hat's only commercial competitors. Mind you, Red Hat is actually earning a profit these days, even though their price::earnings ratio is about 100.
So is it time to invest in Linux stocks (again), except this time with an eye for the long-term instead of the wild ride of the late 90's?
You're confusing propositions with facts. And you're promoting that confusion quite assertively (but not factually).
Moving to Linux is good for geeks and non-geeks
on
Moving To Linux
·
· Score: 2
I've been using Linux as my primary desktop at home since 1998, but I picked up the book primarily for my wife (who made the leap to Linux somewhere around 2001 due to a magical combination of getting tired of having to constantly reboot and the availability of StarOffice). My hope was that she would find Marcel's style approachable enough to dig into the things she was interested in on her own. That didn't pan out, though; she's happier just asking me a quick question when she runs into a dead end.
However, I found myself learning a few things about KDE from the book that I wasn't aware of due to my WindowMaker / Gnome past:
Kooka is a good front end for scanning, a little nicer than xsane
K3B is a kick-butt CD burner, miles ahead of xcdroast: look ma, I can burn DVDs! (This is the one part of the book my wife refers to occasionally)
KDE's printing configuration console is well thought out and works well
So I went from being a primarily command-line oriented guy to trusting a little more in the nice GUI apps that KDE supplies... and it has been good. I'm happy that I picked up Marcel's book.
By the way, he didn't write this book using the French chef schtick that he uses for his Linux Journal "Cooking with Linux" columns, for which I'm quite thankful. I enjoy the columns, but a full book of that would be too much. Instead, he adopts a personal tone that is straightforward and pleasant to read.
Google is too smart for you. "Silent links" (invisible links) are discarded by the search engine, which actually does analyze CSS to determine what would be visible to a regular human being, and how emphasized that text is. Invisible text = thrown out (or if you're really bad, means that your pages will be removed from the Google index entirely).
This BusinessWeek article states that you'll be able to work disconnected, then sync up the next time you connect. So IBM is building replication capabilities into their products. Makes sense; IBM has replication know-how from both their Lotus Notes and their DB2 database products.
Yes it does. Unfortunately the HiSi function for the Neuros is only supported under Windows, so I haven't tried it out. (Yup, staunch Linux/OGG person here.)
Still, posts on the Neuros forums (like this one) suggest that the feature works rather well. So you get unlimited uses of the ID feature for a $200 player--if you think you'll want to identify 200 songs over the course of a couple of years, that's like getting the Neuros player for free. Good deal.
Tendonitis learns you good, fast
on
Two-Fisted Computing
·
· Score: 4, Interesting
I'm a righty, but I switched to a left-handed mouse about five years ago after a one-week motorcycle trip around the Great Lakes (intense vibration) followed by a one-week click-fest through the original Fallout RPG.
Because I make my living as a computer geek, I was surprised and dismayed to find just how messed up my right wrist was after that boneheaded combination of events. I tentatively switched over to using the mouse left-handed, meaning to do it temporarily, but discovered that within a week I was pretty comfortable, and within two weeks I had fully adjusted.
Five years later I'm still using it left-handed, which seems to mess up both righties and lefties when they try to use my workstation. I used to use xwrits to remind me to take regular mouse breaks so I don't (&%# up this wrist too, but I've been bad lately.
I've thought about a two-mouse system, but editing code and writing tech docs really lends itself to a keyboard. If I was an artist maybe a two-mouse system would let me switch brushes and colours midstroke, but I'm no artist. Just a recently minted ambidextrous person. I suppose if I were in Russia, I could say that the mouse manipulated ME...
Just an internet hiccup, it seems to be working fine again. I can't imagine why birdsandbeans.ca would be censored by the USA; "Darn pinkos with their environment- and third-world friendly coffee!" doesn't strike me as a phrase that ISPs would be muttering:)
Slashdot Mirror
Wow... I could have written what you wrote, that hits really damned close to home. Two differences:
* I find that occasionally my dreams do produce practical solutions to programming problems, occasionally forcing me to just get up and try it out - because otherwise I'll toss and turn and obsess over it.
* I no longer get into the zone much anymore, as I have two toddlers that dislodge me very regularly (for better or for worse... mostly for better on the human side).
So, your story was only one year after the earliest story of this type that I could find on Slashdot:
http://tech.slashdot.org/story/01/09/23/1614214/SirCam-on-Linux-via-WINE
2001... wow.
Yes, really. I think its fair to say that this particular set of libraries (almost the entire set of libraries in the state of Georgia) has staked its future on an open-source library system: they're staking their future not just on the code they've developed and released to open source, but also on Linux, Apache, PostgreSQL, Perl, SpiderMonkey, Mozilla/XUL, Jabber, lib-dbi / lib-dbd, and umpteen Perl modules. Take away the OSS pieces that make up that system, and you take away their ability to function as a library. So yes, their future within the state of Georgia depends on OSS.
Of course, the nice thing with an open source ILS is that there is one problem that cannot be solved with money: if your vendor happens to go out of business, or decides to stop supporting your product, or decides to hike your fees immensely and you don't have a good means of exporting all of your data (including serials and acquisitions) to another system, then you're basically screwed. And given all of the consolidation and private equity investments going on in the library system vendor world at the moment, I'm not particularly confident about the viability of any of the current vendors. Open source at least gives you the assurance that you are not locked in with a given product; you own your data, and you can extend or enhance the system if you have the skills in-house or are willing to fund the skills.
And what makes more sense: having one hundred libraries paying $10K - $25K or more per year in basic support fees to vendors, or investing those same funds (or equivalent resources) towards a common library system that can be shared freely? Acquisitions and serials are tricky, sure, but you can make a lot of headway with that kind of common resource available. Hey, look: one institution has already joined GPLS to help develop an acquisitions module for Evergreen. And yes, the open source solution requires support resources: but proprietary solutions require support personnel as well... who, at least in the case of our library system, are each required to take a $3K training course to be certified as a system administrator, and have to take another $3K training course if they want to touch the API that the vendor has made available for the system.
This is true, as far as the bibliographic information goes. There are lots of open-source packages for working with MARC records, like pymarc (Python) or File_MARC (PHP). But the rest of the system is proprietary: holdings records, (which copies do you hold, in which locations, and where is that copy currently - loaned out, lost, on reserve, etc), circulation records, user records, acquisitions records. Sure, it's all just a database schema mapping exercise, if your vendor's license allows you to touch that data directly. Sadly, the past generation of libraries seems to have accepted vendor lock-in as a matter of course; a mistake that we're paying for now and which led directly to the development of Evergreen.
Wow. This is just so wrong that I don't know where to begin. First, Voyager is far from the market leader (in either usable interfaces or in market share). See Second, the underlying database doesn't mean a thing if you aren't given the APIs to actually modify or extend your primary application, unless you're willing to reimplement the entire application -- in which case, why bother paying for a library system in the first place. And in most cases, when the vendor has made an API available, you have to pay extra fee per potential developer to receive the documentation and to be eligible for paid support for their API (which, of course, is an additional support fee over and above your standard support fees). Third, most librarians I know couldn't care less about what technology their system is built on. They're focused on providing the best possible service to their users. Over the past few years, the library community has started to realize that there are some pretty cool Web interfaces out there in the wild that their vendors aren't providing for us. So we've been going through exercises like NCSU's use of Endeca (on the proprietary side) and Koha, Evergreen, and WPopac (on the open-source side) to try and correct the situation. Librarians rock, you know.Sure. Check out OSS4LIB for a list of different open-source library systems. On the smaller end of the scale, OpenBiblio, PhpMyLibrary, and Emilda get mentioned a fair bit.
Umm, dude -- you need to ask for help in the right forum. There are a few of us trying to nail down the installation instructions right now.
It's a lot more complex than just running rpm against a tarball. This application depends on Apache, PostgreSQL, a Jabber server, libmemcached, CVS versions of SpiderMonkey and the lib-dbi / lib-dbd packages, and a host of Perl modules. Georgia's installation runs on top of 25 servers. Getting a secure, stable system up and running was understandably the Evergreen team's first priority; making the source and as much documentation available as they already have was a courtesy that they didn't even have to extend. However, they've gone further than that: they have made a commitment to open source, and librarians-who-do-development / developers-for-libraries thank them for that!
The project is still in its infancy as far as growing outside of Georgia. That's why a few of us are starting by trying to build Open-ILS (aka Evergreen) on our own, following the existing installation instructions. We've already been able to tease out a few more details in the process of trying to create distribution-specific instructions... not that the developers are trying to keep installation hard, it's just that they've lived and breathed this for the last year and a half and so some details are second-nature to them.
I have been quite impressed with the Evergreen team's technical capabilities and commitment to open source. They have contributed patches to the upstream libdbi and libdbd packages as a matter of course, and they pull in a ton of Perl modules that are either not packaged in most distributions, or on which they have dependencies on newer versions than are available in current distros. The team is committed to a transparent open source model comparable to the Linux or PostgreSQL development models and have posted a first draft of that model for comment.
So, the first hurdle is to set up an Evergreen environment outside of Georgia. I've come very close with Ubuntu and Gentoo, but have one more hurdle to clear. All the way along, the developers have been extremely responsive to my questions.
Once we nail down the "gotchas", my personal goal is to create a VMWare image that can be easily redistributed for demonstration purposes. I agree that packaging is obviously going to be an important aspect of the project, but it naturally has to follow a well-documented manual install process.
Evergreen is going to spread outside of Georgia, and it's going to spread fast. The University of Windsor has already announced their intention to work with Evergreen to build an acquisitions system.
Hmm. Now I come back, view the same long deeply nested thread, and it's all neat and pretty, just like the deities intended.
Either the Slashdot devs were working hard on correcting problems on their side, or IE7 is a little bit shaky. And we all know that Cmdr Taco et al are members of the Cult of Slack, and that Microsoft is pure evil, so...
Poor Slashdot. You guys go to all of the trouble of switching to clean tags + CSS, and Microsoft goes and updates their browser so that nested levels of comments render completely unintelligibly with text splashed laughably about.
I suspect Microsoft fears you, Slashdot hordes. Although you would have thought that the IE7 developers might have tested how Slashdot renders before releasing an update that is sure to incite fear and loathing in the masses gathered here (just like witches at black masses -- oh lord yeah!).
True, but most Linux distributions give you the option of installing compatibility libraries, which would solve your problem nicely.
One good reason to use DB2 is that neither MySQL nor PostgreSQL can scale to the same size of database as DB2. Using DB2's "Distributed Partitioning Facility", or DPF, with DB2 Enterprise Server Edition you can partition data over hundreds of machines using a shared-nothing approach. This means you can create a data warehouse of a hundred terabytes that 1) appears to your applications as just a single database 2) retrieves query results blazing fast (this is the "shared nothing" part; each machine knows what data it is responsible for, so a query that comes in to the database is automatically distributed to the responsble machines, which go off and do their work, then return the results to a single co-ordinator that pipes them back to the application -- it's like RAID-0 for databases).
It's certainly possible to create a multi-terabyte database using MySQL or PostgreSQL -- but you're going to be waiting way the hell longer to get a result back from your queries.
Reason 2: You get support from IBM. IBM lives and dies by its enterprise level support -- 24x7 customer service around the world. Sure, you can purchase support from MySQL AB or from some PostgreSQL shop, but it's not going to approach the level of service that you'll get from IBM. In fact, to get a level of support comparable to what IBM offers for DB2 Express at $4874 for the first year and $1218/year for support after the first year, you would need to buy MySQL Network Gold for $2995/year. So after two years, going with DB2 Express will actually cost you less than half of what MySQL support will cost!
Really? I thought you said you hadn't installed any other databases... anyways, installing DB2 is a matter of:
./db2setup
[click through GUI installer:
- enter password for instance user
- enter password for fenced procedure user
- enter password for database administration server user
]cat ".
source
db2start
There's a DB2 HOWTO at the Linux Documentation Project that provides detailed instructions for many popular Linux distributions, including special prereqs or workarounds for distros like Gentoo.
IBM VisualAge C++ is also supported on Linux.
You might ask "why?" -- and the answer is the same as if you asked "Why are companies still selling Fortran compilers?": a whackload of existing, working code and a build environment that would cost mega-bucks to port to a new set of tools.
It would be interesting to see how VisualAge compares on ChaoticCoyote's benchmark. I hope IBM contacts him to get into the next edition of his benchmarks.
Oh, and provide your rationale to those (like Rusty and Dan) who actually set the standards. Whining about it on Slashdot is hardly the way to achieve any change.
Here's what the FHS says about /media, by the way:
Given that the FHS is headed up by notables Rusty Russell and Dan Quinlan, I've got a lot of confidence in their judgement.
Can anyone track down the original Gartner report that indicated 50% of server sales would be Linux by 2008? The linked article just mentions the Gartner report (and all-important statistic) in passing, but doesn't provide a proper reference for fact-checkers. Google didn't do the trick for me, it did turn up an article about an IDC report released in June 2004 that predicted Linux server shipments would rise to 29% in 2004, a fairly significant difference.
Novell's stock is looking pretty attractive at $5.80, given that they're trading close to their 52-week low and now own SuSE, one of Red Hat's only commercial competitors. Mind you, Red Hat is actually earning a profit these days, even though their price::earnings ratio is about 100.
So is it time to invest in Linux stocks (again), except this time with an eye for the long-term instead of the wild ride of the late 90's?
You're confusing propositions with facts. And you're promoting that confusion quite assertively (but not factually).
I've been using Linux as my primary desktop at home since 1998, but I picked up the book primarily for my wife (who made the leap to Linux somewhere around 2001 due to a magical combination of getting tired of having to constantly reboot and the availability of StarOffice). My hope was that she would find Marcel's style approachable enough to dig into the things she was interested in on her own. That didn't pan out, though; she's happier just asking me a quick question when she runs into a dead end.
However, I found myself learning a few things about KDE from the book that I wasn't aware of due to my WindowMaker / Gnome past:
So I went from being a primarily command-line oriented guy to trusting a little more in the nice GUI apps that KDE supplies... and it has been good. I'm happy that I picked up Marcel's book.
By the way, he didn't write this book using the French chef schtick that he uses for his Linux Journal "Cooking with Linux" columns, for which I'm quite thankful. I enjoy the columns, but a full book of that would be too much. Instead, he adopts a personal tone that is straightforward and pleasant to read.
Google is too smart for you. "Silent links" (invisible links) are discarded by the search engine, which actually does analyze CSS to determine what would be visible to a regular human being, and how emphasized that text is. Invisible text = thrown out (or if you're really bad, means that your pages will be removed from the Google index entirely).
This BusinessWeek article states that you'll be able to work disconnected, then sync up the next time you connect. So IBM is building replication capabilities into their products. Makes sense; IBM has replication know-how from both their Lotus Notes and their DB2 database products.
Still, posts on the Neuros forums (like this one) suggest that the feature works rather well. So you get unlimited uses of the ID feature for a $200 player--if you think you'll want to identify 200 songs over the course of a couple of years, that's like getting the Neuros player for free. Good deal.
I'm a righty, but I switched to a left-handed mouse about five years ago after a one-week motorcycle trip around the Great Lakes (intense vibration) followed by a one-week click-fest through the original Fallout RPG.
Because I make my living as a computer geek, I was surprised and dismayed to find just how messed up my right wrist was after that boneheaded combination of events. I tentatively switched over to using the mouse left-handed, meaning to do it temporarily, but discovered that within a week I was pretty comfortable, and within two weeks I had fully adjusted.
Five years later I'm still using it left-handed, which seems to mess up both righties and lefties when they try to use my workstation. I used to use xwrits to remind me to take regular mouse breaks so I don't (&%# up this wrist too, but I've been bad lately.
I've thought about a two-mouse system, but editing code and writing tech docs really lends itself to a keyboard. If I was an artist maybe a two-mouse system would let me switch brushes and colours midstroke, but I'm no artist. Just a recently minted ambidextrous person. I suppose if I were in Russia, I could say that the mouse manipulated ME...
Just an internet hiccup, it seems to be working fine again. I can't imagine why birdsandbeans.ca would be censored by the USA; "Darn pinkos with their environment- and third-world friendly coffee!" doesn't strike me as a phrase that ISPs would be muttering :)