Young Scientist ot Year - ya wha ?

daveirl

This post has been deleted.

daveirl

This post has been deleted.

Meh

instead of simply tinkering with existing code he went down to the socket layer and reworked it at the protocol level

I'm still skeptical...more buzzwords and technobabble from anonymous sources, but still a suspicious lack of hard information. After all, this is the Young Scientist award, and I thought "peer review" was an essential part of science.

It will be interesting to see how he managed to "rework the http protocol" to make it 5x faster without violating any of the various HTTP/TCP/IP RFCs...

dahamsta

below are two interesting slashdot.org posts in response to the news of the innovative new software:

Here's a few more...

* Maybe he's a big fan of whitespace?
* "Warning! Your internet connection is not optomized!" -- Someone finally clicked that damn pop-up.
* I have a theory - if he feeds the processor the right series of commands, he could create a resonance cascade that would create a bubble in spacetime which would cause all time inside it to pass at a faster pace than normal. The downside on this would be that you age four times as fast while surfing, but hey, there's always something ...
* And I bet he has an Uncle Rael who just helped create the first human clone, too, right?

As Slashdot said.. what happened to Sarah Flannery's 'revolutionary' encryption algorithm.

As has been pointed out in several places - including the main index for the /. post - Flannery's system shouldn't be used as a comparison, because it was a valid system that got bad press because it was discovered after the Young Scientist competition that it was open to attack; by William Whyte, Michael Purser and.... Sarah herself. See John Young's Cryptome for more:

http://cryptome.org/flannery-cp.htm

[ NOTE: This isn't the same as saying that Adnan's system isn't valid too. I'm just saying that comparing to Sarah's system is wrong, since ultimately hers was; it was just vulnerable to attack, and the crypto community can still learn from it. It remains to be seen what will happen with Adnan's. I await with bated breath. ]

adam

Sev

What did he write it with?

longword

Originally posted by Meh
It will be interesting to see how he managed to "rework the http protocol" to make it 5x faster without violating any of the various HTTP/TCP/IP RFCs...

You don't have to go rewriting RFCs or w3c specs. You could do what Apple have done with Safari - claimed to be over 300% faster than Internet Explorer.

Meh

Originally posted by longword
You don't have to go rewriting RFCs or w3c specs.

But that's exactly what he claims to have done:

he went down to the socket layer and reworked it at the protocol level

zynaps

So according to this anecdotal evidence and checking out, the browser simply is more efficient than (presumably) IE's browser, by a factor of 4, or whatever.

That's nice, but does anyone really think web browsers are the bottleneck in internet access? :P

I mean, if IE takes even as much as 20 milliseconds (pfft!) to display a page that has taken 4000 milliseconds to load over a dialup connection, what good is reducing the rendering time to 4ms?

What's sad about it is the press starts going on about 400% faster internet thanks to some genius student, and then everyone who reads newspapers, watches tv or talks to people is echoing the misled claim.

ACH :P

zynaps

sceptre

Just a quick (possibly irrelevant) post:

The DVD player. Where has this come from? The reason I'm bringing this up is that the DVD consortium exert a strong (silly in my view but there you go) control on authorised DVD players, both hardware and software. To get access to a CSS code, they insist on a non-refundable deposit of one million dollars.

Assuming that the guy hasn't tossed a million smackers on the table, the DVD player obviously either comes from somewhere else (a licensed player tossed in) or is using hacked codes. Not that I care about the hacked code possibility but it either has fallen foul of patent law here or will fall foul of copyright law if we ever get the evil Euro-DMCA transcribed into Irish law. Either possibility would fall foul of this as presumably using a licensed DVD player would be outside the license granted (it's the same problem the Linux community are facing - who's going to come up with a million smackers to licence a player for distribution that can't go totally open source anyway)

Like I say - it might be irrelevant. It would have to be licensed properly before "the package" could become available even if Karlin's source is bang on.

On that note though, the whole "new browser" thing is a bit of a red herring if he's actually done something different - can you blame people for concluding the whole thing was a new table-top cold fusion with vagueness like that (from all sides, not just the media)? Sarah Flannery was far more open about what she'd done - the pending patent is a non-issue if it's already been applied for and any comments about a lack of public or peer review are entirely valid.

daveirl

This post has been deleted.

rob1891

In fairness to Sarah Flannery she amended her paper for the European competition (which she won along with a fair number of other awards) and all the work is available publicly (as it must be in science?). I had a look at the paper, and the maths looks rather stunning for a 16 year old, I would definitely have to get out my college books and (re)learn a fair bit before I'd be able to grasp any of it.

I don't think it's in the spirit of science to not publish algorithms and methods used in enough detail so other scientists can make the same experiments(implementations) and verify the results (aka peer review as has been said).

I also don't think an application is a suitable project for the competition .... at least what does a dvd playing browser have to do with Computer Science. A concept on how to improve speeds and then a sample implementation would be what I'd expect, then again that would be boring ...

I guess he's on the ball then

I hope anyway that he publishes some technical info and has indeed got the browser speed up, +200% in rendering would still be some achievement for someone of his age, and his claims, or the claims that the media might have encouraged him to make should be let pass, he's only a wee impressionable lad!

rob

jmcc

Originally posted by Meh
I'm still skeptical...more buzzwords and technobabble from anonymous sources, but still a suspicious lack of hard information. After all, this is the Young Scientist award, and I thought "peer review" was an essential part of science.

I have to agree on this one. There is a major lack of any hard technical details. As for the MLE source, I think that most of those guys have their heads so far up their asses that they don't know whether to sh1t or shave.

The important part in Karlin's report (and the most important detail to emerge so far) is that the browser renders the page faster than other browsers but does not speed up download. Those UCD people are going to look like major idiots who thought that a faster rendering engine meant that the browser was downloading faster. Still though if it is a more efficient rendering engine, and it is his own work, he deserves the win.

Regards...jmcc

daveirl

This post has been deleted.

daveirl

This post has been deleted.

daveirl

This post has been deleted.

acous

Still though if it is a more efficient rendering engine, and it is his own work, he deserves the win.

Is it really possible that someone who writes a html rendering engine so much faster than any other browser to come out with statements about its speed like "At seven times it actually crashes so I have limited it to six.". And what he said on RTE about dvd quality video streaming not being possible until now. People talking bullshít like that just pisses me off, no matter how good his socket code might be.

sutty

I have to say alot of it seams to media word spindling and some by him-self as well. Like the streaming DVD, the amount of bandwidth taken to do this would be in the Higher DSL packiges. Also could I ask one very important thing...

WHY DO YOU WANT TO WATCH A DVD WHILE YOUR LOOKING AT WEB PAGES!!! I mean you will miss most of the freaken movie ffs.

I for one dont think that this project should have won seeing as he does not give any of the tech info for anything behind the program or any of the "science" be hind it. I also dont like the way he is being called a web wiz when he doesn't even have anything on the web. You would think that he would have a web page or something about his project somewhere... wouldn't you?

eth0_

Just re people saying why did he win when his project isn't science...isn't the the YS officially a science and technology competition? Sort of like saying computer science isn't science

sutty

Aye it is, but he should at least give a discription of the science used in his project, or how he did it. Seeing as he has it copyrighted in all. Like he doesn't give any detales of how he dont anything.

LizardKing

This guy's story is gone global

I pity the poor bastad , if it turns out to be a bag o' shíte then he's gonna get some stick for some time...

if in turn it does turn out to be "The Real Deal" then he's gonna be extremely rich....

Just to start a few conspiracies :eek: heres a few ideas why he has not released demos or a web presence.

- He is being paid off by Microsoft to halt the release so they can copy his technology so they do not lose ground in browser wars

- He is being paid off by Netscape to halt the release so they can copy his technology so they regain browser market

- He is being paid off by the telecoms industry to halt release so that they can continue to sell ISDN / DSL

- He has actually not written anything and only presented mozilla with plugins

last ones probably closest to truth but as I'm optimistic I hope he turns out to have done something decent though :rolleyes:

longword

Originally posted by Meh

You don't have to go rewriting RFCs or w3c specs.

But that's exactly what he claims to have done:

he went down to the socket layer and reworked it at the protocol level

No, that's not what Adnan claims. That's your interpretation of a 3rd hand report from a journalist who talked to a researcher who may possibly have spoken to Adnan. My interpretation of that same 3rd hand report is that it's a brand new browser that uses its own implementation of the HTTP protocol as opposed to reworking the core of another browser core like Gecko or KHTML.

I'm still waiting for the XWEBS web site to appear. If he's churning out code at 1,500 lines a day, one page of HTML shouldn't be too hard to muster...

spod

Originally posted by jmcc
... As for the MLE source, I think that most of those guys have their heads so far up their asses that they don't know whether to sh1t or shave.

I take it you haven't left Waterford and wandered around the MLE campus lately then? Nice, lovely, smart people who are refreshingly down to earth, have good coffee, yummy bagels and killer chairs. Oh and they're mainly cynical Europeans with only a few happy clappy merikans brought over from the parent medialab.

jmcc

Originally posted by spod
I take it you haven't left Waterford and wandered around the MLE campus lately then? Nice, lovely, smart people who are refreshingly down to earth, have good coffee, yummy bagels and killer chairs. Oh and they're mainly cynical Europeans with only a few happy clappy merikans brought over from the parent medialab.

Me, in Dublin? I have decided that Waterford, being the first city of Ireland, is now the capital.

Ok so rather than knowing whether to sh1t or shave, they spend time drinking coffee and eating bagels. Though they may be very nice people.

I still think it was a waste of good money. It could have been spent on providing broadband. After all, just how many Starbucks clones is enough?

Regards...jmcc

Overlord

hes a bluffer

sceptre

Karlin has another update on her blog

Further info on Adnan Osmani's faster browser: I spoke to Dick Ahlstrom today, the Irish Times science editor who wrote the Times piece on Adnan Osmani and his browser project for the Young Scientist Exhibition. Dick gave permission to reprint his story (the Times site is subscription only) for free here.

Dick has some programming background and an extensive science background. He said that of course the judges were concerned that the browser might appear to be doing what it was, in truth, not doing, so they brought in three additional judges,all from senior technology positions, to evaluate it. In addition they sent it the University College Dublin's well-respected computer science department, where researchers there put it through its paces, running it continuously for two days. It did exactly what Adnan said it could do, flawlessly, the UCD people told Dick when he contacted them. In addition, they would have been able to examine it at code level. (Bernie Goldbach also has a posting here about the browser from a programmer, Fergal Byrne, who saw it running and talked to Adnan.)

Dick stressed that Adnan was not trumpeting this project at all. He warned him that he'd probably get intense media scrutiny, perhaps as intense as Irish schoolgirl Sarah Flannery did when she won the same contest with her encryption algorithm. Interestingly, Dick pointed out that this story, too, had been of modest interest until the London Times stuck it on page 1 and made a big deal out of Sarah being 1) a girl and 2)young. Then everyone picked it up and the hype began.

Personally, I don't know how kids so young deal with the voracious public appetite for every detail about them, as well as all the criticism and scrutiny, when basically they're just doing something they really like and (in the case of both Adnan and Sarah) not deliberately searching for a huge profile. While Adnan's project may or may not end up rewriting browser history, and may or may not bring him real rather than fleeting fame, it's important to remember that he's a teenager and simply submitted a project that has, so far, been given as good a going-over as professional computer scientists could give it in the weeks leading up to the competition awards night.

The last paragraph is certainly worth noting.

phaxx

Indeed, but I won't believe it no matter who tells me it works, I'll want to see a good explanation for myself.

GUI

at this stage for his own sake..
hope it is the real deal,
or he will be a name we never forget for the wrong reasons :-)

monument

Originally posted by jmcc
just how many Starbucks clones is enough?

better then a Starbucks on every street

Dempsey

He'll be in the the same bracket as BILL GATES:D .

I'd put my money on that he used the two most simplest words in the computer programming world,.........COPY.........PASTE.... with a big "I'm going to scam the science awards head on him"

longword

Another tiny trickle of information thanks to Karlin...

http://radio.weblogs.com/0103966/

Looks like most of those 1.5 million lines of code were indeed written by Borland or Microsoft. But it does at least sound like he wrote something that might be worth a prize.