The company I work for, Garlik has two products that are run off semantic web technology. DataPatrol (for pay) and QDOS (free, in beta).
We use RDF stores instead of databases in some places as they are very good at representing graph structures, which are a real pain to real with in SQL. You often hear the "what can RDF do that SQL can't" type arguments, which are all just nonsense. What can SQL do that a field database, or a bunch of flat files can't? It's all about what you can do easily enough that you will be bothered to do it.
A fully normalised SQL database has many of the attributes of an RDF store, but
a) when was the last time you saw one in production use?
b) how much of a pain was it to write big queries with outer joins?
RDF + SPARQL makes that kind of thing trivial, and has other fringe side benefits (better standardisation, data portability) that you don't get with SQL.
I guess it shouldn't be a surprise to see the comments consisting of the usual round of more-or-less irrelevant jokes and snide commentary - this is Slashdot after all - but I can't help responding.
I'm certainly knee deep in RDF and OWL; I develop a GPL'd RDF engine called 3store which is moving OWL-wards in the near future.
I've been a low-level DB wrangler for some years, and FWIW I now find semantic-web structures much easier to deal with than SQL. The simple fact that you dont/need/ a schema to assert data and make queries is hugely useful, not to mention the inference RDFS can do, which is pretty lame my AI stnadards, but is very fast,and still quite powerful.
For "professional" applications there is JACK (http://jackit.sf.net/) that allows applications to route audio between each other and the soundcard. The majority of the good apps for linux audio support this.
It requires apps to run thier audio thread with SCHED_FIFO scheduling though, so its not really ideal for simple mp3 players.
- The live audio stream to be broadcast on Friday and Thursday (probably between ~ 2 P.M. and 9 P.M. on both days) is available at these LiveIce servers:
x http://plugin.org.uk:2300/liveice (currently set to max. 50 clients) x http://politik.uni-duisburg.de:2300/liveice (max. 20 clients)
As posted to the linux-audio-developers mailing list.
Generally pitch scaling of this kind is done with a "Phase Vocoder", as in my Linux audio plugin, pitch_scale (http://plugin.org.uk/).
In order to run it in real time you have to limit the oversampling of the window function, to maybe 16 times, and choose a tradeoff between time accuracy (the smeary sound you hear in PV pitch shifted stuff) and pitch accuracy (the FM like pitch wobble you hear in low rate MP3s).
However, if you dont need to produce it in realtime you can throw a few weeks of computing power at it and produce something nice sounding, espeically if you window for each input sample.
Ceres 3 (for linux and IRIX, http://www.music.columbia.edu/~stanko/About_Ceres3.html) can produce very good results, but I dont think it handles very large files well, as it has to pre process them, and you wont have the patince to wait for it to do a whole symphony.
You don't understand the problem, the echo comes from the signal being sent to the other end, being picked up on the mic at the other end and sent back.
The delay is in the order of 1500ms.
You need a big (~100k point) adaptive FIR to cancel it, getting a PC to do it in realtime is challenging, and there won't be many resources left to do other things. We have a gentner at work, and its is fine. If you need >2 channels it would be cheaper than the eqivalent PC hardware, and its much easier to set up.
There are a number of very promising pro-level applications for linux, whihx I won't bother listing as there are several good webpages that will do that.
I will say however that most of them are not ready for everyday use yet, most are alpha and still under heavy development.
On the hardware side, both M-Audio Deltas (24bit analogue, up to 10 channels) and RME Hammerfalls (digital only, 3 ADAT ports each way + SPDIF, my personal choice) work well and are supported. Both are available from any decent electronic music shop.
If you are a DSP programmer then pitch in and write some LADSPA plugins (see http://www.ladspa.org/), the API is very simple and well documented adnd lots of applications can use them.
'Xtian' is not derogatory, at least not originally, the X is a greek letter (Chi?) not the roman X, and was a very common abbreviation at one time, used by Chritians themselves.
Fundie is a different matter, but doesn't apply to Christians in particular, you get Linux fundies too;)
There is a company in the UK (in the vicinity of hampshire) that has an implementation of the sphere idea, they don't do any head tracking but they do project the image onto the outside of the sphere. The sphere is floated on a cushion of air (noisy but low friction). I have tried it and it works, though it is a bit unresponsive.
I did know what they were called, but I have forgotten, and am very drunk, so I follow up with thier details on Monday if anyone wants to know.
Slashdot Mirror
The company I work for, Garlik has two products that are run off semantic web technology. DataPatrol (for pay) and QDOS (free, in beta).
We use RDF stores instead of databases in some places as they are very good at representing graph structures, which are a real pain to real with in SQL. You often hear the "what can RDF do that SQL can't" type arguments, which are all just nonsense. What can SQL do that a field database, or a bunch of flat files can't? It's all about what you can do easily enough that you will be bothered to do it.
A fully normalised SQL database has many of the attributes of an RDF store, but
a) when was the last time you saw one in production use?
b) how much of a pain was it to write big queries with outer joins?
RDF + SPARQL makes that kind of thing trivial, and has other fringe side benefits (better standardisation, data portability) that you don't get with SQL.
I guess it shouldn't be a surprise to see the comments consisting of the usual round of more-or-less irrelevant jokes and snide commentary - this is Slashdot after all - but I can't help responding.
How do you limit wifi to 1st and bussiness class customers?
Though, without inseat power in cattle class its not as much use.
I'm certainly knee deep in RDF and OWL; I develop a GPL'd RDF engine called 3store which is moving OWL-wards in the near future.
/need/ a schema to assert data and make queries is hugely useful, not to mention the inference RDFS can do, which is pretty lame my AI stnadards, but is very fast,and still quite powerful.
I've been a low-level DB wrangler for some years, and FWIW I now find semantic-web structures much easier to deal with than SQL. The simple fact that you dont
For "professional" applications there is JACK (http://jackit.sf.net/) that allows applications to route audio between each other and the soundcard. The majority of the good apps for linux audio support this.
It requires apps to run thier audio thread with SCHED_FIFO scheduling though, so its not really ideal for simple mp3 players.
Errr... you must be thinking of something else, ardour doesn't have a phase vocoder.
FWIW I've also supplied a few patches to ardour and have written several phase vocoders for Linux.
- Steve
- The live audio stream to be broadcast on Friday and Thursday (probably
between ~ 2 P.M. and 9 P.M. on both days) is available at these LiveIce
servers:
x http://plugin.org.uk:2300/liveice (currently set to max. 50 clients)
x http://politik.uni-duisburg.de:2300/liveice (max. 20 clients)
As posted to the linux-audio-developers mailing list.
Generally pitch scaling of this kind is done with a "Phase Vocoder", as in my Linux audio plugin, pitch_scale (http://plugin.org.uk/).
3 .html) can produce very good results, but I dont think it handles very large files well, as it has to pre process them, and you wont have the patince to wait for it to do a whole symphony.
In order to run it in real time you have to limit the oversampling of the window function, to maybe 16 times, and choose a tradeoff between time accuracy (the smeary sound you hear in PV pitch shifted stuff) and pitch accuracy (the FM like pitch wobble you hear in low rate MP3s).
However, if you dont need to produce it in realtime you can throw a few weeks of computing power at it and produce something nice sounding, espeically if you window for each input sample.
Ceres 3 (for linux and IRIX, http://www.music.columbia.edu/~stanko/About_Ceres
I think you answered your own question.
I don't know what variety of "english" you speak, but where I come from (England) the pronunciation of
"thought" bears no relation to "hot".
You don't understand the problem, the echo comes from the signal being sent to the other end, being picked up on the mic at the other end and sent back.
The delay is in the order of 1500ms.
You need a big (~100k point) adaptive FIR to cancel it, getting a PC to do it in realtime is challenging, and there won't be many resources left to do other things. We have a gentner at work, and its is fine. If you need >2 channels it would be cheaper than the eqivalent PC hardware, and its much easier to set up.
There are a number of very promising pro-level applications for linux, whihx I won't bother listing as there are several good webpages that will do that.
I will say however that most of them are not ready for everyday use yet, most are alpha and still under heavy development.
On the hardware side, both M-Audio Deltas (24bit analogue, up to 10 channels) and RME Hammerfalls (digital only, 3 ADAT ports each way + SPDIF, my personal choice) work well and are supported. Both are available from any decent electronic music shop.
If you are a DSP programmer then pitch in and write some LADSPA plugins (see http://www.ladspa.org/), the API is very simple and well documented adnd lots of applications can use them.
- Steve
Other problems (when I tried it - a couple of years ago now).
* Noise
* Lag
* Wobble (the sphere wasn't that rigid)
* Image alignment problems
Also if I remember right there was a problem with direction and turning, but I can't think what it was.
That said it was quite impressive.
'Xtian' is not derogatory, at least not originally, the X is a greek letter (Chi?) not the roman X, and was a very common abbreviation at one time, used by Chritians themselves.
;)
Fundie is a different matter, but doesn't apply to Christians in particular, you get Linux fundies too
Nope, its handrolled php3.
- Steve
What!
I mean I dislike Microsofts software as much as the next man, but suggesting they orchestrated this is just plain paranoid.
Yeesh.
There is a company in the UK (in the vicinity of hampshire) that has an implementation of the sphere idea, they don't do any head tracking but they do project the image onto the outside of the sphere. The sphere is floated on a cushion of air (noisy but low friction). I have tried it and it works, though it is a bit unresponsive.
I did know what they were called, but I have forgotten, and am very drunk, so I follow up with thier details on Monday if anyone wants to know.
Wierdly enough, thats exactly what we though when we wrote it, we were gaggin for a cease and desist, but we never got one...
- Darth Steve
They came form forbidden planet in the UK, I think.
- Steve (steve@totl.net)
Sounds a bit like Robert Rankin to me, though Rankin is a technophobe, very very British and uses a lot of occult "in jokes".
Anyone read both?