Ah! That Old Misconception again...
on
Shirky On P2P
·
· Score: 1
If only I had a fiver for every time somebody has posted to some forum saying words to the effect of "If only the net had been designed *properly* from the start - we'd never have to implement all these kluges!"
It's precicely BECAUSE of the absence of a over-arching design that the net has been and will continue to be so successful. I agree that's a mite hard to grasp when people are used to devoting their lives to designing elegant systems (and/or are attracted to the M$ concept of centralised control), but I believe it to be true.
In design terms the net is a singularity - it needs to have a degree of chaos at its heart to function.
I've been noticing this ever since the rise of mass Internet use - the tendency for people to assume that what they read on the web is current, which in turn is aggrevated by the fact that so much content on the net is undated.
I recently complained to Clay Shirky that none of his essays were dated, thereby making it very diffcult to work out whether he was talking about things in the light of certain events or not. I suppose most readers assumed he'd written them that day or something, when in fact several were over five years old.
The UK GSM Association today blesses M Services - a mobile services spec written by OpenWave and BT Genie which includes WAP 1.2 and 2.0 stuff plus OpenWave extensions for presentation layer and download.
Why one earth is this thread turning into an anti-WAP ranting session? WAP will be *enhanced* under through M Services, not killed off.
The M Services initiative is about instituting the standards on handsents that WAP hasn't had before.
> In today's world, the reality is that very few users don't accept cookies.
Exactly. All the people posting here going "cookies don't work because people turn them off" are on planet Slashdot.
HOWEVER - one thing that's not been mentioned yet is that if you use mod_usertrack to cookie your logs and you get a user who does not accept a cookie, it creates a stream of unique IDs - one for every request that user makes.
So - those people who turn their cookies off and go to Apache servers could be looking like hundreds or even thousands of users! Hooray!
> could a few folks who are running unique
> cookies on a large site count 'em and count IP
> addresses in the same period and give us their
> ratio?
Certainly. We were counting users by IP address/browser type combination for about a year on three sites getting between 5,000 and 50,000 users per month based on that calculation. We then decided to use cookies (expiring at 5 years). We saw an increase in user numbers of about 20% for each site.
Which was nice, seeing as we then went for ABCe audits and got lots of advertising money!
>Mark Thomas would not get the audience and the
>platform to speak from if he did not play round
>and do silly stuff like this.
Much as I hate to think that this is the case, in fact, Mark Thomas probably gets the audience and the platform *because* what he does is ultimately ineffective.
If he was really making a difference, there's no way he'd get TV (or any other) airtime. Don't get me wrong - I love his stuff, but it has to fall short of actually hitting the spot.
I agree. The Web was a much better place when failure was always a distinct possiblity. Failure creates jobs, failure keeps me in business, failure is very good for everybody.
Information wants to be free, and much content actively *defies* categorisation, since "to define is to limit" (as the Buddhists say).
That's your problem when trying to impose a top-down categorisation system. The answer, IMHO is to look at the problem on a more granular level: self-categorisation through probability, for example (e.g. like Autonomy do).
G
IMHO all this "OSX is really UNIX" business totally misses the point. There are 18 gazillion falvours of *NIX already - who the hell wants another one?
Instead, we should regard OSX as a continuation of the great Mac tradition of being different. Stop treating it like it's some long-lost cousin coming back to the fold and look at it like we look at BeOS, etc. New, different, probably riddled with bugs but in the end, wonderful.
Most of the larger search engines have Usenet search options. The trouble is, none are as yet as big as Dejanews, which, it has to be said, has done a good job of indexing the vast amount of data on Usenet. In the past, that is:-(
On second thoughts, I think the BSD license would be a better proposition.
It also depends on the nature and complexity of the code as well a great deal. Still, perhaps it's something to bear in mind when the lawyers start to bite months (and mucho cash) out of a potential project.
One area where open-sourcing might be a business advantage is in code ownership diputes.
I work for an agency that takes on large web projects for clients. The contract we get the clients to sign says that we own all the copyright on the code we create for them. If they want to own it we can license it to them for a fee.
Most clients are happy with this, since they can't maintain the code themselves. They simply commission us to make changes to is as part of the project's maintanance.
However, from time to time a client won't sign until we let them own the code. This is bad from our point of view because we want to be able to re-use the code on other projects, etc. This is also bad because it can delay the start of a project while the lawyers bicker over the whole thing.
Now, we've not actually done this yet, but we could, I suppose, suggest to the client that wants to own the code that we make it open source. That way, we all "own" the code, and if they want to use it under the terms of the GPL then fine, 'cos we love the GPL too. We might also hope that making it open source might actually make it better.
I've not really thought about this idea very much so there might be some fatal flaw in there somewhere, but so far it seems to make sense to me.
For what it's worth, we *really* wanted to install and use WebSphere on Solaris with Oracle, but in the end gave up - we just coudn't install it. Their help desk didn't seem to be able to support the proudct under Solaris, and we just got fed up and used WebLogic instead.
Not only were prices the same as on the highstreet, but they had no Nike products. That's like a baker not selling sliced white!
Basically, boo.com had so many things wrong with it that I'm only surprised it lasted this long.
Yes, they regarded good design as being synonymous with the use of Flash, knew nothing about usability, and probably little about the complexity of their fulfilment problems, but at the base of all that was simple, old-fashioned inability to compete. I quite liked the ads though - but they were utterly awful at pulling in the punters.
G
PS: Talking of Flash - will www.moonfruit.com be next for a well-deserved diet prior to closure?
One thing that sometimes gets forgotten is that you can reduce calls to the server by using client-side strategies. For example, putting the raw results of a database query in a Javascript array then mainpulating it (sorting, etc.) in the browser can be very fast and groovy. It's not a substitute for server-side caching, but if you can take any load onto the client then it's good.
Power issues for mobile devices will prove to be a real rain on the their parade.
I've read what seems like 10,000 articles in the popular (and even technical) press that predict mobile computing will be super huge in 3 to 5 years time. Here in Europe, we'll have GPRS and 2Mb/s on our handheld devices by 2002 (apparently), so journos are predicting we'll have full-motion video feeds, etc., on our cellphones and the like.
Well, we might, but we'll also have to develop full-on biceps to carry the batteries that will have to power the things.
I can confirm that Mr Ungodly-Tomas is indeed a complete fool.
He mailed me pretending to be a young student wanting to "get into" animal rights and hacking *weeks* after his psudonym and mailhost details were blown away by NTK, one of the most widely-read mailing lists in the UK.
Interesting. There seems to be two separate parts to this: the idea that ISPs could charge based on traffic, and the second is the sharing of your source code with the ISP.
Looking at the first idea, I don't know much about US hosting but I assume they're on a flat-rate model - yes? Over here in the UK, most ISPs that I know of in fact charge by bandwidth used in exactly the way you describe. Most review the charge on a monthly or even weekly basis. For example, the host we're on the moment says that if we go above 1G per week they charge us extra (the idea being that we negotiate with them when they blow the whistle, it's not simly an automatic increase). So, they have every incentive to make sure our sites are available and popular. By US standards, however, the charge is high (about $1,500 a year below 1G a week).
On the second issue of sharing code, I'm unclear as to whether you mean that it's connected to the first point or not, but if you intend it to be an extra incentive to the ISP to be nice to you (a payment in kind if you like) then perhaps it would be attractive to smaller ISPs who couldn't afford enterprise licences for commercial shopping cart applications etc. But why would they want to enter into a restricted licence with you when there are anyway several GPL realeases of all the applications you mention? Everybody likes to think their wheel is better invented than anyone else's, but the ultimate peer review is the GPL. I don't think you can have your cake an eat it, which is what you're implying here.
Slashdot Mirror
If only I had a fiver for every time somebody has posted to some forum saying words to the effect of "If only the net had been designed *properly* from the start - we'd never have to implement all these kluges!"
It's precicely BECAUSE of the absence of a over-arching design that the net has been and will continue to be so successful. I agree that's a mite hard to grasp when people are used to devoting their lives to designing elegant systems (and/or are attracted to the M$ concept of centralised control), but I believe it to be true.
In design terms the net is a singularity - it needs to have a degree of chaos at its heart to function.
I've been noticing this ever since the rise of mass Internet use - the tendency for people to assume that what they read on the web is current, which in turn is aggrevated by the fact that so much content on the net is undated.
I recently complained to Clay Shirky that none of his essays were dated, thereby making it very diffcult to work out whether he was talking about things in the light of certain events or not. I suppose most readers assumed he'd written them that day or something, when in fact several were over five years old.
Ho hum. (Hows that for OT?)
-----BEGIN PGP MESSAGE-----
c i+ 3108yJzdnV0AhM
m zD 7ctN3XLHRP2yej
h 5X bJeo6Fb+3dVEgX
v pk B3VMf5eUbt0cpa
i E4 3YNVuYeafCS4Xn
9 CA CFdBE6qJW9LlA6
/ Dc SFmbvc7uE+YqJY
X yg JlVrzh3jf6tivK
p 23 7ycLzHeBUfYgmp
k ig CCCYGVwDqJzz2p
S Fw FOSJrh9MckR/wT
U oA Cf2IMhppXcklSN
7 Dw BCL/MzErLru+mY
d Ws x3D8t2lLyMNPPI
6 L5 aqvoNyf7iBgwPZ
Version: PGPfreeware 6.0.2i
qANQR1DBwU4DHoNwP2q9kEMQCACxSm5gUmsAZSeiru1YZqh
B47QgAxn3Cb2gYuRphKbpqkAg+rTBXCIoHwnafL0bNPPoWB
jH2b2WRp7iUP7s/pScXzUB/KVJe1bxBQL2zqJglESR4QH6B
cBokCTv8gK77yqzb4SHV83WtsIcLe8Y7b8wEHuYPqyt2uR+
B5X9P78EbF2vIIB5sxAfHlopZaGGa2CMkJa+HdDxc6RrIG2
ZMeORMub5h31BR0rmBi34PvoDp2/r6/3v/e3ZIsd7tizAEt
wBY86ZVjkeg0wP5hrJohd354TQPAN2AD2nY3yDlQpSHG99e
bsimyOZn1xBjNHl1NLtSMn9z5x7JdBO2m4xZt01kqOJAL0l
huzY9dj5KpdfiAUn37ZMo3du2J6ZPCAesnYVs9ylTP9KWmi
BzjIhBb93DHhkWfxiwn7cYmQ9pnjAJcXHTXQe7T9IDu/MsT
J/eSDrrMjcvM/5DS6Muw6lgKG8hQlWFCXRqAY61giZlqycJ
bEZ4PXcDybvsjAjM8gdS/xDGR3/XUh8OFZ9S4AnR13UWtnH
80V7HlRg8ggslyIi6hnjgIR/LoNxEbdWIC+MtceoO8ekE/2
W5NOIoMyhnsARYS4acwx9eC2f449DSTT9HlHABDmFhstpCn
0rVKL+fyLBc16Bld4wD6ZbU/b+7vJwQbMz15Rue0unH6LqS
EKNq
=//mi
-----END PGP MESSAGE-----
Are you using the latest version? I'm not getting the bahaviour you discribe.
The UK GSM Association today blesses M Services - a mobile services spec written by OpenWave and BT Genie which includes WAP 1.2 and 2.0 stuff plus OpenWave extensions for presentation layer and download.
Why one earth is this thread turning into an anti-WAP ranting session? WAP will be *enhanced* under through M Services, not killed off.
The M Services initiative is about instituting the standards on handsents that WAP hasn't had before.
Exactly. All the people posting here going "cookies don't work because people turn them off" are on planet Slashdot.
HOWEVER - one thing that's not been mentioned yet is that if you use mod_usertrack to cookie your logs and you get a user who does not accept a cookie, it creates a stream of unique IDs - one for every request that user makes.
So - those people who turn their cookies off and go to Apache servers could be looking like hundreds or even thousands of users! Hooray!
G
> could a few folks who are running unique
> cookies on a large site count 'em and count IP
> addresses in the same period and give us their
> ratio?
Certainly. We were counting users by IP address/browser type combination for about a year on three sites getting between 5,000 and 50,000 users per month based on that calculation. We then decided to use cookies (expiring at 5 years). We saw an increase in user numbers of about 20% for each site.
Which was nice, seeing as we then went for ABCe audits and got lots of advertising money!
See - cookies DO work!
G
>The University of Michigan has one of the most
>respected social sciences/statistics departments
>in the world
... who also make up words. What the hell is "religiosity"? Do they also measure fat people's "obesiosity", or how about levels of "faceiousisity"?
Bloody Americans.
>Mark Thomas would not get the audience and the
>platform to speak from if he did not play round
>and do silly stuff like this.
Much as I hate to think that this is the case, in fact, Mark Thomas probably gets the audience and the platform *because* what he does is ultimately ineffective.
If he was really making a difference, there's no way he'd get TV (or any other) airtime. Don't get me wrong - I love his stuff, but it has to fall short of actually hitting the spot.
But how would Tripwire help against the Lion worm? Wouldn't it just rub your nose in the fact you'd just got 0wnd? JJ
NO. Bob Dobbs is the one tru God. Send me 1000 dollars and I'll prove it to ya.
I agree. The Web was a much better place when failure was always a distinct possiblity. Failure creates jobs, failure keeps me in business, failure is very good for everybody.
Information wants to be free, and much content actively *defies* categorisation, since "to define is to limit" (as the Buddhists say). That's your problem when trying to impose a top-down categorisation system. The answer, IMHO is to look at the problem on a more granular level: self-categorisation through probability, for example (e.g. like Autonomy do). G
IMHO all this "OSX is really UNIX" business totally misses the point. There are 18 gazillion falvours of *NIX already - who the hell wants another one?
Instead, we should regard OSX as a continuation of the great Mac tradition of being different. Stop treating it like it's some long-lost cousin coming back to the fold and look at it like we look at BeOS, etc. New, different, probably riddled with bugs but in the end, wonderful.
Most of the larger search engines have Usenet search options. The trouble is, none are as yet as big as Dejanews, which, it has to be said, has done a good job of indexing the vast amount of data on Usenet. In the past, that is :-(
On second thoughts, I think the BSD license would be a better proposition.
It also depends on the nature and complexity of the code as well a great deal. Still, perhaps it's something to bear in mind when the lawyers start to bite months (and mucho cash) out of a potential project.
One area where open-sourcing might be a business advantage is in code ownership diputes.
I work for an agency that takes on large web projects for clients. The contract we get the clients to sign says that we own all the copyright on the code we create for them. If they want to own it we can license it to them for a fee.
Most clients are happy with this, since they can't maintain the code themselves. They simply commission us to make changes to is as part of the project's maintanance.
However, from time to time a client won't sign until we let them own the code. This is bad from our point of view because we want to be able to re-use the code on other projects, etc. This is also bad because it can delay the start of a project while the lawyers bicker over the whole thing.
Now, we've not actually done this yet, but we could, I suppose, suggest to the client that wants to own the code that we make it open source. That way, we all "own" the code, and if they want to use it under the terms of the GPL then fine, 'cos we love the GPL too. We might also hope that making it open source might actually make it better.
I've not really thought about this idea very much so there might be some fatal flaw in there somewhere, but so far it seems to make sense to me.
JJ
For what it's worth, we *really* wanted to install and use WebSphere on Solaris with Oracle, but in the end gave up - we just coudn't install it. Their help desk didn't seem to be able to support the proudct under Solaris, and we just got fed up and used WebLogic instead.
A pity. But there you go.
Not only were prices the same as on the highstreet, but they had no Nike products. That's like a baker not selling sliced white!
Basically, boo.com had so many things wrong with it that I'm only surprised it lasted this long.
Yes, they regarded good design as being synonymous with the use of Flash, knew nothing about usability, and probably little about the complexity of their fulfilment problems, but at the base of all that was simple, old-fashioned inability to compete. I quite liked the ads though - but they were utterly awful at pulling in the punters.
G
PS: Talking of Flash - will www.moonfruit.com be next for a well-deserved diet prior to closure?
Lovely to bask in the glow of CONTROL and POWER, isn't it!
Bill Gates - I can see why he does what he does. It's the buzz.
>Cache your dynamic data
One thing that sometimes gets forgotten is that you can reduce calls to the server by using client-side strategies. For example, putting the raw results of a database query in a Javascript array then mainpulating it (sorting, etc.) in the browser can be very fast and groovy. It's not a substitute for server-side caching, but if you can take any load onto the client then it's good.
JJ
Power issues for mobile devices will prove to be a real rain on the their parade.
I've read what seems like 10,000 articles in the popular (and even technical) press that predict mobile computing will be super huge in 3 to 5 years time. Here in Europe, we'll have GPRS and 2Mb/s on our handheld devices by 2002 (apparently), so journos are predicting we'll have full-motion video feeds, etc., on our cellphones and the like.
Well, we might, but we'll also have to develop full-on biceps to carry the batteries that will have to power the things.
More reserch money for power!
G
I can confirm that Mr Ungodly-Tomas is indeed a complete fool.
He mailed me pretending to be a young student wanting to "get into" animal rights and hacking *weeks* after his psudonym and mailhost details were blown away by NTK, one of the most widely-read mailing lists in the UK.
G
Interesting. There seems to be two separate parts to this: the idea that ISPs could charge based on traffic, and the second is the sharing of your source code with the ISP.
;-)
Looking at the first idea, I don't know much about US hosting but I assume they're on a flat-rate model - yes? Over here in the UK, most ISPs that I know of in fact charge by bandwidth used in exactly the way you describe. Most review the charge on a monthly or even weekly basis. For example, the host we're on the moment says that if we go above 1G per week they charge us extra (the idea being that we negotiate with them when they blow the whistle, it's not simly an automatic increase). So, they have every incentive to make sure our sites are available and popular. By US standards, however, the charge is high (about $1,500 a year below 1G a week).
On the second issue of sharing code, I'm unclear as to whether you mean that it's connected to the first point or not, but if you intend it to be an extra incentive to the ISP to be nice to you (a payment in kind if you like) then perhaps it would be attractive to smaller ISPs who couldn't afford enterprise licences for commercial shopping cart applications etc. But why would they want to enter into a restricted licence with you when there are anyway several GPL realeases of all the applications you mention? Everybody likes to think their wheel is better invented than anyone else's, but the ultimate peer review is the GPL. I don't think you can have your cake an eat it, which is what you're implying here.
Overall, I'd listen to your wife
JJ