That address (208.65.153.253) doesn't work for me (but thanks for the quick response). Does it work for you?
When I do whois and nslookup youtube.com I get that address (208.65.153.253) and the same result as in the grandparent post. This doesn't look like a genuine Youtube record:
Could somebody please post Youtube's real IP address please?
(Cue oblig links to the IP address of a famous Christmas Island domain)
Serialization is the future, but it needn't be XML
on
The Future of XML
·
· Score: 1
Networked computing means more communication of data among heterogenous systems. But XML is just one way to serialize data. Also, we should remember that data is often stored permanently somewhere, not merely transferred and forgetten in a chain of computers handling one transaction. There will continue to be multiple formats at the low level (XML, JSON and language/application specific forms and so on) for some time. If there is to be consolidation to fewer formats, there is no reason to doubt that XML could be squeezed out quite soon.
If we see JSON being used more in the browser, and hence in the webserver, then why should developers waste effort to convert it to and from XML? We could have entire applications using JSON for everything from the database storage to the browser.
Enormous predicted demand leads to huge supply, leads to economies of scale, leads to lower prices. The important thing is the supply, not the demand as such.
The article implies that Linus could kill the GPL3 by not using it for Linux. But most of the software in a (GNU/)Linux distro isn't part of the kernel and thusly will be GPL2 or 3 no matter what Linus says. And let's remember that Linus doesn't have copyright on the whole kernel, far from it in fact.
So GNU/Linux will be GPL for ever, with the only debate being over the timing of the gradual change from v2 to v3.
If Linus did try to start a new project from scratch with another license, then it'd make sense to (I never thought I'd say this!) start using the HURD.
> Now I won't be able to drink it while using MS Word.
Don't worry. I'm sure you could download some crack. And if that doesn't work, you could forget the alcohol altogether and instead download some crack.
No amount of fiddling with calendars will change the fact that there are about 365.25 days in a year.
If we could alter the orbit of the earth or increase the rate at which it turns perhaps we could make 1000 days in a year. Only then would it make sense to try to decimalize time.
There was nothing wrong with the British Rail privatisation. The safety record was no worse than it was before privatisation and the punctuality of trains has got worse after it was renationalised a couple of years ago.
And yes, it was renationalised, I don't want any pedants claiming that Network Rail isn't a state body.
The private owner, Railtrack, managed to take on a 50% increase in passenger numbers and still keep a reasonable safety and punctuality record - although people will argue over how to interpret certain statistics.
The only reason privatisation wasn't more successful was because of the massive interference and regulation by Labour and Tory governments. Why did they try to promote too much competition in railways when the railways are facing massive competition from roads, planes et cetera?
By the way, I don't like making claims like the above without citing some good quality sources. I just feel like ranting before double checking my facts!
All this political bickering is good news for fusion. The more the merrier. It makes it less likely that one country will have a monopoly on the technology.
If the EU go ahead, it will mean the other players have more of an incentive to make it work, instead of the oil industry trying to scupper it.
> They don't seem to feel the need for > journalists to understand basic system > design issues
<sarcasm> Damn journalists. Next they'll be asking basic questions like 'how does the Electoral College work?'. Doesn't everyone know that elections aren't meant to be in public. </sarcasm>
Nothing wrong with saying 95% of people have below average intelligence. If 95% of people had an IQ of exactly 99 and 5% of poeple had an IQ of 120 then the average would be approx. IQ 100. Meaning that 95% of people would have IQ less than 100.
It's unlikely though.
If he said 95% of people have below median intelligence, that would be nonsense
>> Xtreme Programming (that's a damn stupid name, isn't it?)
> YES! YES! YES!
Why don't we call it 'Iterative Programming' or something?
Now, for me to go even more OT - I feel like a rant on software engineering: Each time our team is asked to update an application we are required to write a new requirements document. This document just documents the changes made. This means that instead of a single document (or set of documents) that describes that current system, we have hundreds of documents - none of which describe the system fully and most of which are out of date!
The software company that I work for seems to treat each change as a totally separate application and not as a single evolving system.
Surely the purpose of any document is so that it might be of some use to somebody in the future, an obvious example is a new team member - but when I joined my current team, nobody suggested that I read the requirements documents because they must have known that they were useless. Why don't they take it the obvious next step and ask themselves "How can we write docs that are useful?".
Maybe it's because everyone thinks that good documentation just makes you more replaceable... if every employer thinks like that I'm changing career. I'm studying for a mathematics degree at the same time as working as a programmer. At least in mathematics you never get away with a dodgy proof - I'm sick of dodgy code.
> software engineering will still take place in the western countries > anyone can write code
Why do you think software engineering can be separated so easily from the coding? Surely the best thing is to have one competent person who can do both?
It would make it harder for the coder and the engineer to point fingers at one another.
And anyway, according the most of the estimates I see (to be taken with a pinch of salt of course), coding only takes up 10% of the time - why take all the risks of splitting this off to a different person in a different country.
Also, in my experience, it's only when you start coding that you realise how the design could be improved, and you realise the requirements need clarifying, et cetera. I think this is the main lesson of Xtreme Programming (that's a damn stupid name, isn't it?)
I think that software should be developed by less people, but make sure those people are well educated about everything from testing and requirements gathering all the way to how processors work at a low level.
Slashdot Mirror
It's back working for me now. And I've realised now that it doesn't seem to have been a DNS problem after all - please ignore my posts on this thread!
That address (208.65.153.253) doesn't work for me (but thanks for the quick response). Does it work for you?
When I do whois and nslookup youtube.com I get that address (208.65.153.253) and the same result as in the grandparent post. This doesn't look like a genuine Youtube record:
...Server Name: YOUTUBE.COM.ZZZZZ.GET.LAID.AT.WWW.SWINGINGCOMMUNITY.COM
...Server Name: YOUTUBE.COM.IS.N0T.AS.1337.AS.WWW.GULLI.COM
I'm in Ireland, not Pakistan, so I should be able to see something! Am I getting an IP routing problem, or just poisoned DNS?
Could somebody please post Youtube's real IP address please?
(Cue oblig links to the IP address of a famous Christmas Island domain)
Networked computing means more communication of data among heterogenous systems. But XML is just one way to serialize data. Also, we should remember that data is often stored permanently somewhere, not merely transferred and forgetten in a chain of computers handling one transaction. There will continue to be multiple formats at the low level (XML, JSON and language/application specific forms and so on) for some time. If there is to be consolidation to fewer formats, there is no reason to doubt that XML could be squeezed out quite soon.
If we see JSON being used more in the browser, and hence in the webserver, then why should developers waste effort to convert it to and from XML? We could have entire applications using JSON for everything from the database storage to the browser.
> A real Unix geek would use a symbolic link.
That's what my boss said when I claimed out that I can't be in two places at once.
Enormous predicted demand leads to huge supply, leads to economies of scale, leads to lower prices. The important thing is the supply, not the demand as such.
Space On a Shoestring?
Aren't they going to need something stronger than shoestring for a space elevator?
The article implies that Linus could kill the GPL3 by not using it for Linux. But most of the software in a (GNU/)Linux distro isn't part of the kernel and thusly will be GPL2 or 3 no matter what Linus says. And let's remember that Linus doesn't have copyright on the whole kernel, far from it in fact.
So GNU/Linux will be GPL for ever, with the only debate being over the timing of the gradual change from v2 to v3.
If Linus did try to start a new project from scratch with another license, then it'd make sense to (I never thought I'd say this!) start using the HURD.
*ducks*
> Meanwhile I switched from Slackware to Gentoo this week and nobody seems to notice.
Of course we haven't noticed - it hasn't compiled yet.
What's your IP address?
Bring on the Cultural Revolution!
> Now I won't be able to drink it while using MS Word.
Don't worry. I'm sure you could download some crack.
And if that doesn't work, you could forget the alcohol altogether and instead download some crack.
> start writing a coherent stance
.. or just rip off a fellow slashdotter's comment.
> From now on, life is rated AO
ESRB has ruled however, that this ruling doesn't apply to slashdotters because they are rarely invited for 'hot coffee'.
That's wrong too.
It should be:
popup.innerHTML += '<div><a href="' + array[i] + '"></a></div>';
I hate wasting time,
Wasting valuable surfing time by doing work, that is.
It would be really useful if it could tell where the crime was about to occur .. in advance
Your "wife" had a Y chromosome?
Very disconcerting to find that out on your wedding night, I'm sure.
No amount of fiddling with calendars will change the fact that there are about 365.25 days in a year.
If we could alter the orbit of the earth or increase the rate at which it turns perhaps we could make 1000 days in a year. Only then would it make sense to try to decimalize time.
There was nothing wrong with the British Rail privatisation. The safety record was no worse than it was before privatisation and the punctuality of trains has got worse after it was renationalised a couple of years ago.
And yes, it was renationalised, I don't want any pedants claiming that Network Rail isn't a state body.
The private owner, Railtrack, managed to take on a 50% increase in passenger numbers and still keep a reasonable safety and punctuality record - although people will argue over how to interpret certain statistics.
The only reason privatisation wasn't more successful was because of the massive interference and regulation by Labour and Tory governments. Why did they try to promote too much competition in railways when the railways are facing massive competition from roads, planes et cetera?
By the way, I don't like making claims like the above without citing some good quality sources. I just feel like ranting before double checking my facts!
All this political bickering is good news for fusion. The more the merrier. It makes it less likely that one country will have a monopoly on the technology.
If the EU go ahead, it will mean the other players have more of an incentive to make it work, instead of the oil industry trying to scupper it.
> They don't seem to feel the need for
> journalists to understand basic system
> design issues
<sarcasm>
Damn journalists. Next they'll be asking basic questions like 'how does the Electoral College work?'. Doesn't everyone know that elections aren't meant to be in public.
</sarcasm>
Nothing wrong with saying 95% of people have below average intelligence.
If 95% of people had an IQ of exactly 99 and 5% of poeple had an IQ of 120 then the average would be approx. IQ 100.
Meaning that 95% of people would have IQ less than 100.
It's unlikely though.
If he said 95% of people have below median intelligence, that would be nonsense
>> Xtreme Programming (that's a damn stupid name, isn't it?)
... if every employer thinks like that I'm changing career. I'm studying for a mathematics degree at the same time as working as a programmer. At least in mathematics you never get away with a dodgy proof - I'm sick of dodgy code.
> YES! YES! YES!
Why don't we call it 'Iterative Programming' or something?
Now, for me to go even more OT - I feel like a rant on software engineering: Each time our team is asked to update an application we are required to write a new requirements document. This document just documents the changes made. This means that instead of a single document (or set of documents) that describes that current system, we have hundreds of documents - none of which describe the system fully and most of which are out of date!
The software company that I work for seems to treat each change as a totally separate application and not as a single evolving system.
Surely the purpose of any document is so that it might be of some use to somebody in the future, an obvious example is a new team member - but when I joined my current team, nobody suggested that I read the requirements documents because they must have known that they were useless. Why don't they take it the obvious next step and ask themselves "How can we write docs that are useful?".
Maybe it's because everyone thinks that good documentation just makes you more replaceable
> software engineering will still take place in the western countries
> anyone can write code
Why do you think software engineering can be separated so easily from the coding? Surely the best thing is to have one competent person who can do both?
It would make it harder for the coder and the engineer to point fingers at one another.
And anyway, according the most of the estimates I see (to be taken with a pinch of salt of course), coding only takes up 10% of the time - why take all the risks of splitting this off to a different person in a different country.
Also, in my experience, it's only when you start coding that you realise how the design could be improved, and you realise the requirements need clarifying, et cetera. I think this is the main lesson of Xtreme Programming (that's a damn stupid name, isn't it?)
I think that software should be developed by less people, but make sure those people are well educated about everything from testing and requirements gathering all the way to how processors work at a low level.