...Speaking of wasted time, I discovered this morning that the SGI Origin
that I've been programming on wouldn't "turn over" (read: boot). The
system diagnostics (which produce output roughly akin to, well, I suppose
what one would expect from a $30k machine) pointed at the scsi controller
and commented "I think you may want to have a look at that." SGI is
sending us overnight a new 'logic carrier' which the technician described
obliquely as containing "the scsi controller, as well as just about
everything else." He also advised us to "tow it home."
Re:Not Sun and Sony, but *Apple* and Sony
on
The Faded Sun
·
· Score: 1
Without entertainment, which provided 30 percent of the company's revenue and nearly all its profits, Sony...
From this Wired article (posted on Slashdot last month) which is worth a read, because it discusses some of the issues you raise. You may be right about music, but everything else is going like gangbusters (especially movies).
Re:Not Sun and Sony, but *Apple* and Sony
on
The Faded Sun
·
· Score: 2, Informative
For Sony, it's the entertainment division
Hello? The entertainment division of sony is the only part that's profitable.
Here's an article by a Mr. Ziemann (cited in the Snyder piece) that attempts to make the argument I was making in my second paragraph. I think his data is interesting but doesn't support his conclusions. In part (and I'm playing devil's advocate a bit here), this is because just a drop in number of new releases doesn't necessarily mean that the recording industry is not providing interesting material; the vast majority of music released is not "interesting" to the extent that it generates large commercial sales. Record companies would argue that there simply focusing more on what's profitable (as they have to do to satisfy their stockholders). As I said above, I think we need to weight the releases by their "popularity" through some other measure (for instance, concert tickets) because this will adjust for the disparity in album releases from popular and unpopular acts.
At the same time, I believe that it's precisely the fact that record companies are releasing fewer new albums that leads me to other means to acquire new music.
The article is a good read, though, at least because he uses lots of expletives in an attempt to justify his conclusions.
This is cool and schmool, but where exactly are the collaborating applications that need to share and lock resources across Internet?
Any web service which needs to conduct a database transaction will potentially need to lock resources. It may be that the application is structured so that a request is a single transaction, but more complicated application may require multiple interactions, thus a longer transaction and the need for locking.
For instance, take a web-services application that allows you to edit an, I don't know, address book entry. You retrieve the address book entry in one request, and store the edited values in another. Now, if another instance of the application on another machine comes along and retreives and stores the same address book entry between your request and your store, then when you store, the previous edits are lost. Hence the need for locking. This is obviously a simple, contrived example. Believe me, I've laid awake at night because of the difficulty of distributed transactions.
Why doesn't someone do a study that correlates the number of CDs sold year over year to the number of new releases year over year weighted by the popularity of those releases (via some other measure, like concert ticket sales, etc.)? This might provide evidence to prove or disprove the common statement that record sales are hurt by the lack of fresh material more than by file sharing.
I'm tired of people (on both sides of the music sharing issue) trotting out the same rhetorical arguments all the time. There's nothing new here. Snyder's arguments about growth in other media don't prove his point; rather they show that the decline in record sales can't be blamed on the slowing economy. I believe I bought fewer CDs from major labels this year, not because I was downloading them from the internet, but because I wasn't interested in them. My money instead goes to new, independant artists (via CD Baby a fantastic way to buy music that doesn't make you feel like a criminal.)
I for one don't believe that we're going to (or that we should) do away with copyright. I just believe that the current music industry is going to be end-runned. I hope that we don't end up legislating it's continued existence. By that I don't mean that we should permit illegal file sharing, but that we shouldn't mandate a technological solution which only allows big media to create content.
I can see this point. I think the analogy is faulty. If you liken DNA to a computer program, then consider a single organism to be a run of that program. A single organism can crash just like a run of a program can.
Now there are certainly programming methodologies modeled on evolution. But that's not what he's talking about. What he's talking about is using pattern recognition to reduce errors in computer programs, I assume, although he doesn't say this, by making them more tolerant of a range of inputs. Evolution has nothing to do with pattern recognition, other than that both are stochastic processes. Evolution is tolerant of environmental pressure by being massively parallel (to borrow another CS metaphor). And even then it's sometimes overwhelmed (think ice age). His programs would be more tolerant of errors because they used better algorithms (namely pattern recognition).
I think it's a bullshit analogy. As I said before, I'm not sure if this analogy is key to his argument, but I don't give him a lot of cred for it.
If you think about it, if you make a small change to a program, it can result in an enormous change in what the program does. If nature worked that way, the universe would crash all the time. Certainly there wouldn't be any evolution or life.
<cough>Bullshit.</cough>
This guy obviously knows nothing about biology. A single base change in DNA can result in mutations that cause death or spontaneous abortion. As little as a change in a single 'character' can be lethal. That's a pretty "small change" that results in a pretty big "crash."
I'm not sure if this invalidates his argument, but it certainly doesn't do much for his credibility.
Re:The client doesn't control the Future of Java
on
The Future of Java?
·
· Score: 1
It's used a lot for server-side applications,
Now I may sound like an idiot for saying this, but java never made sense as a server-side language. It wasn't designed as a server-side language, it was designed as a client-side language. I've never really understood why people say "we use java on the server because it's write-once run-anywhere." It doesn't make sense: on a server you have complete control of the hardware so compiling to machine code is both possible and good for performance. Also, the things that make cross platform coding really difficult are more easily abstracted in a server environment than in a client environment (in my opinion, no one, possibly save mozilla, has truely created a cross-platform client app).
There have to be other reasons why people use java on the server side. I believe it's because of the library. I think the library is what makes java popular, and it's the library that makes it so much easier to code server-side stuff in java. The library, and other tools like app servers do a good job of abstraction on the server-side. Other languages lose in this regard, particularly C++, which doesn't even have a hashtable (hash_map<> isn't part of the standard STL).
Perhaps more importantly, there are quite a few bugs in the GUI libraries (both Swing and AWT) that make it difficult to write highly polished applications.
The vast majority of the highly polished client apps written in java seem to be java editors (Together, IntelliJ, Eclipse, etc.). And even these aren't exactly models of cross platform goodness. This is because it's exteremely difficult to write good cross platform client apps in java. Much more difficult than writing cross platform server apps in java. There is really only one GUI toolkit (AWT/Swing) for java (SWT is only linux and win, and the linux support isn't that good). And this toolkit is difficult to use and even more difficult to extend. (Try writing your own widget in Swing, believe me you don't want to.) That said, I believe java gets closer to a cross platform client library than anything else (again, possibly save mozilla which I think could be a contender but it's hard to know given how fucking enormous in scope it is).
All this is to say that I hope that the inclusion of java on windows motivates Sun or someone to really focus on improving Swing so that it's:
Possible to write really good client side apps that behave appropriately on the various platforms. Start with supporting more of the common widgets well (like sortable tables, &c).
Easier to write said apps (like Cocoa easy).
Possible to extend swing in a fashion that doesn't make you want to defenstrate yourself.
By capitalizing on java's strength (namely, the library) and making that library best of class for GUIs, sun can win in the client space. But does sun care?
I'd love to see a service like Citysearch (which posts, amoung other things, listings of bands playing in your local area, as long as your local area is somewhat metropolitan), but that actually linked to samples of the band's music, and had the ability to quickly listen to a bunch of samples for bands playing tonight. You could do a ton of tie-ins: buy CDs, buy tickets, buy shirts, send mail to the band, etc.
Mp3.com, EmergentMusic.com, others: lots of work.
on
Discovering New Music?
·
· Score: 3, Informative
Slashdot did a similar story a while back and someone posted
a number of alternative sources for new music. Here's another one: EmergentMusic.com. Some pretty good stuff on there; lots of crap too, but that's to be expected.
In general, I find using any of these sites, including and especially MP3.com, to find decent music you haven't heard elsewhere is a ton of work. I continue to contend that, when someone comes up with a decent business model, services which suggest new music will make more money than companies which sell music. I have yet to see one that I think actually works. This is a difficult task: music is an intensely personal decision and relying on suggestions of others takes trust. I really wonder what ever happend to firefly, and the host of other sites that purported to use neural networks to suggest music based on you rating stuff you know you liked. I think these agent based approaches are more likely to succeed than something like a web log or zine.
The solution to this is making the satellites closer, ie LEO (Low Earth Orbit). Alas, the only potential provider for LEO IP access has died on the vine.
Sigh.
Still waiting for decent, interactive internet access anywhere on the globe.
Or use the timer (like when taking a group photo).
Home Inventory Tracking (Re:Car independant)
on
Step 2, Groceries
·
· Score: 1
My solution to this problem was to write a little app that allows me to record what I have and when it entered the system.
I'm constantly in the grocery story asking myself something like, "Do I need to buy mayonaise?"
Doesn't it seem like grocery stores could offer this as a service? They already track your purchases with the damn discount swipey card. Now they just have to put a Web or SOAP interface on that database so you can pull the data (and integrate it with purchases from other stores). You can know when you last bought something. You could get really slick and have terminals in the store. "Hmm... when did I last buy mayonaise? Let's see."
Of course there are issues:
Security- Do I care if someone else knows when I last bought mayonaise? Probably not. Do I care if someone else knows when I last bought condoms? Probably. (Or rather how (very) long ago I last bought condoms, but anyway.)
Utility- This kind of inventory tracking by purchase is only so useful. Things are not always used at a constant rate. For fresh items you could also track by date of spoilage. See below for another idea.
A better system might combine this with a barcode reader over the trash can. Scan the UPC code when you thow away the package. This doesn't work for vegetables, obviously, but it's a start. Combine this with RFIDs and you're golden. Again, except for vegetables, until we invent some seriously genetically modified veggies.
Where I've seen this done it's been by weight. You scan the item and then put in a bag on a scale. The computer makes sure that the total weight of the bag equals the total of the (known in a database) weights of the items you've purchased. It gets confused, for example, when items are touching something outside the scale. It's not a fool proof technology obviously, but with the guy there watching I'd say it's probably pretty successful.
The "music distribution site" could allow users to post "reviews" and give point ratings to individual bands/songs. A fairly powerful mechanism for locating and suggesting music that you will probably like could be made by informing the site which bands you already know you like, and then you can be given "try this" selections based on bands that were given high ratings by users who also rated your selections high.
This is somewhat like what firefly purported to do. Every time you were presented with a band you could rate that band, and it would use that to recommend other bands based on other users recommendations. It was quite cool but it dissappeared. I'm not sure what happened to it.
The value that major labels provide is simple: exposure. They have the ears: yours. The problem with this is it's cyclical: no one wants to buy CDs from a band nobody's heard of, the only way a band can get heard of is to get radio play, no unheard of band gets radio play.
I would pay good money for a service that consistently recommended decent new music. I'm willing to set the bar fairly low too. It seems like all the services (remember firefly? the original musicnet that mailed out CDs? I can't even find links for these people anymore) that purported to do this went out of business before I could even try to give them money. Contrast this with the MP3.com experience which, although you may find great stuff, it's a ton of work to do. This is an inherently personal service and trust has a lot to do with it.
There are a few record companies out there, such as Aware Records (John Mayer, Train), and Windham Hill that consistently produce good new music I like, such that I'm willing bet on their compilation CDs. There are a few radio stations (like 105.9 the X) that have decent new music shows I'm willing to listen to.
The bottom line is that something like this has got to happen but it's a tricky problem and no one has yet found the right business model. Please, come up with one, I'll line up to send you money.
All these companies from microsoft to phillips keep talking about having a "revolution in handheld computing." Where the hell is it? Can I buy it today? No. We're still 5 years behind where we were when we lost the newton. Grr... I wish these people would put their chips up and let me vote with my wallet.
Slashdot Mirror
You haven't seen Synergy yet, have you?
...Speaking of wasted time, I discovered this morning that the SGI Origin that I've been programming on wouldn't "turn over" (read: boot). The system diagnostics (which produce output roughly akin to, well, I suppose what one would expect from a $30k machine) pointed at the scsi controller and commented "I think you may want to have a look at that." SGI is sending us overnight a new 'logic carrier' which the technician described obliquely as containing "the scsi controller, as well as just about everything else." He also advised us to "tow it home."
From this Wired article (posted on Slashdot last month) which is worth a read, because it discusses some of the issues you raise. You may be right about music, but everything else is going like gangbusters (especially movies).
Hello? The entertainment division of sony is the only part that's profitable.
At the same time, I believe that it's precisely the fact that record companies are releasing fewer new albums that leads me to other means to acquire new music.
The article is a good read, though, at least because he uses lots of expletives in an attempt to justify his conclusions.
-c
Any web service which needs to conduct a database transaction will potentially need to lock resources. It may be that the application is structured so that a request is a single transaction, but more complicated application may require multiple interactions, thus a longer transaction and the need for locking.
For instance, take a web-services application that allows you to edit an, I don't know, address book entry. You retrieve the address book entry in one request, and store the edited values in another. Now, if another instance of the application on another machine comes along and retreives and stores the same address book entry between your request and your store, then when you store, the previous edits are lost. Hence the need for locking. This is obviously a simple, contrived example. Believe me, I've laid awake at night because of the difficulty of distributed transactions.
-c
I'm tired of people (on both sides of the music sharing issue) trotting out the same rhetorical arguments all the time. There's nothing new here. Snyder's arguments about growth in other media don't prove his point; rather they show that the decline in record sales can't be blamed on the slowing economy. I believe I bought fewer CDs from major labels this year, not because I was downloading them from the internet, but because I wasn't interested in them. My money instead goes to new, independant artists (via CD Baby a fantastic way to buy music that doesn't make you feel like a criminal.)
I for one don't believe that we're going to (or that we should) do away with copyright. I just believe that the current music industry is going to be end-runned. I hope that we don't end up legislating it's continued existence. By that I don't mean that we should permit illegal file sharing, but that we shouldn't mandate a technological solution which only allows big media to create content.
-c
Wow! what a fantastic story line. If you don't want it, can I have it?
-c
Yes, obviously you can't read. He said "evolution or life" doesn't crash. I was saying it does.
Now there are certainly programming methodologies modeled on evolution. But that's not what he's talking about. What he's talking about is using pattern recognition to reduce errors in computer programs, I assume, although he doesn't say this, by making them more tolerant of a range of inputs. Evolution has nothing to do with pattern recognition, other than that both are stochastic processes. Evolution is tolerant of environmental pressure by being massively parallel (to borrow another CS metaphor). And even then it's sometimes overwhelmed (think ice age). His programs would be more tolerant of errors because they used better algorithms (namely pattern recognition).
I think it's a bullshit analogy. As I said before, I'm not sure if this analogy is key to his argument, but I don't give him a lot of cred for it.
<cough>Bullshit.</cough>
This guy obviously knows nothing about biology. A single base change in DNA can result in mutations that cause death or spontaneous abortion. As little as a change in a single 'character' can be lethal. That's a pretty "small change" that results in a pretty big "crash."
I'm not sure if this invalidates his argument, but it certainly doesn't do much for his credibility.
Now I may sound like an idiot for saying this, but java never made sense as a server-side language. It wasn't designed as a server-side language, it was designed as a client-side language. I've never really understood why people say "we use java on the server because it's write-once run-anywhere." It doesn't make sense: on a server you have complete control of the hardware so compiling to machine code is both possible and good for performance. Also, the things that make cross platform coding really difficult are more easily abstracted in a server environment than in a client environment (in my opinion, no one, possibly save mozilla, has truely created a cross-platform client app).
There have to be other reasons why people use java on the server side. I believe it's because of the library. I think the library is what makes java popular, and it's the library that makes it so much easier to code server-side stuff in java. The library, and other tools like app servers do a good job of abstraction on the server-side. Other languages lose in this regard, particularly C++, which doesn't even have a hashtable (hash_map<> isn't part of the standard STL).
Perhaps more importantly, there are quite a few bugs in the GUI libraries (both Swing and AWT) that make it difficult to write highly polished applications.
The vast majority of the highly polished client apps written in java seem to be java editors (Together, IntelliJ, Eclipse, etc.). And even these aren't exactly models of cross platform goodness. This is because it's exteremely difficult to write good cross platform client apps in java. Much more difficult than writing cross platform server apps in java. There is really only one GUI toolkit (AWT/Swing) for java (SWT is only linux and win, and the linux support isn't that good). And this toolkit is difficult to use and even more difficult to extend. (Try writing your own widget in Swing, believe me you don't want to.) That said, I believe java gets closer to a cross platform client library than anything else (again, possibly save mozilla which I think could be a contender but it's hard to know given how fucking enormous in scope it is).
All this is to say that I hope that the inclusion of java on windows motivates Sun or someone to really focus on improving Swing so that it's:
- Possible to write really good client side apps that behave appropriately on the various platforms. Start with supporting more of the common widgets well (like sortable tables, &c).
- Easier to write said apps (like Cocoa easy).
- Possible to extend swing in a fashion that doesn't make you want to defenstrate yourself.
By capitalizing on java's strength (namely, the library) and making that library best of class for GUIs, sun can win in the client space. But does sun care?I'd love to see a service like Citysearch (which posts, amoung other things, listings of bands playing in your local area, as long as your local area is somewhat metropolitan), but that actually linked to samples of the band's music, and had the ability to quickly listen to a bunch of samples for bands playing tonight. You could do a ton of tie-ins: buy CDs, buy tickets, buy shirts, send mail to the band, etc.
In general, I find using any of these sites, including and especially MP3.com, to find decent music you haven't heard elsewhere is a ton of work. I continue to contend that, when someone comes up with a decent business model, services which suggest new music will make more money than companies which sell music. I have yet to see one that I think actually works. This is a difficult task: music is an intensely personal decision and relying on suggestions of others takes trust. I really wonder what ever happend to firefly, and the host of other sites that purported to use neural networks to suggest music based on you rating stuff you know you liked. I think these agent based approaches are more likely to succeed than something like a web log or zine.
Sigh.
Still waiting for decent, interactive internet access anywhere on the globe.
-c
Or use the timer (like when taking a group photo).
I'm constantly in the grocery story asking myself something like, "Do I need to buy mayonaise?"
Doesn't it seem like grocery stores could offer this as a service? They already track your purchases with the damn discount swipey card. Now they just have to put a Web or SOAP interface on that database so you can pull the data (and integrate it with purchases from other stores). You can know when you last bought something. You could get really slick and have terminals in the store. "Hmm... when did I last buy mayonaise? Let's see."
Of course there are issues:
- Security- Do I care if someone else knows when I last bought mayonaise? Probably not. Do I care if someone else knows when I last bought condoms? Probably. (Or rather how (very) long ago I last bought condoms, but anyway.)
- Utility- This kind of inventory tracking by purchase is only so useful. Things are not always used at a constant rate. For fresh items you could also track by date of spoilage. See below for another idea.
A better system might combine this with a barcode reader over the trash can. Scan the UPC code when you thow away the package. This doesn't work for vegetables, obviously, but it's a start. Combine this with RFIDs and you're golden. Again, except for vegetables, until we invent some seriously genetically modified veggies.-c
Where I've seen this done it's been by weight. You scan the item and then put in a bag on a scale. The computer makes sure that the total weight of the bag equals the total of the (known in a database) weights of the items you've purchased. It gets confused, for example, when items are touching something outside the scale. It's not a fool proof technology obviously, but with the guy there watching I'd say it's probably pretty successful.
-c
This is somewhat like what firefly purported to do. Every time you were presented with a band you could rate that band, and it would use that to recommend other bands based on other users recommendations. It was quite cool but it dissappeared. I'm not sure what happened to it.
I would pay good money for a service that consistently recommended decent new music. I'm willing to set the bar fairly low too. It seems like all the services (remember firefly? the original musicnet that mailed out CDs? I can't even find links for these people anymore) that purported to do this went out of business before I could even try to give them money. Contrast this with the MP3.com experience which, although you may find great stuff, it's a ton of work to do. This is an inherently personal service and trust has a lot to do with it.
There are a few record companies out there, such as Aware Records (John Mayer, Train), and Windham Hill that consistently produce good new music I like, such that I'm willing bet on their compilation CDs. There are a few radio stations (like 105.9 the X) that have decent new music shows I'm willing to listen to.
The bottom line is that something like this has got to happen but it's a tricky problem and no one has yet found the right business model. Please, come up with one, I'll line up to send you money.
> What do you know, they have the entire article
> online. Enjoy.
Thanks very much.
-c
> I read a really interesting article in the
> Atlantic Monthly Journal, not too long ago.
I'd like to read this. Can you post a reference?
-c
All these companies from microsoft to phillips keep talking about having a "revolution in handheld computing." Where the hell is it? Can I buy it today? No. We're still 5 years behind where we were when we lost the newton. Grr... I wish these people would put their chips up and let me vote with my wallet.