| This 39 message thread spans 2 pages: < < 39 ( 1  ) || |
|A Dipsie tidbit|
Some of us have openly wondered if Dipsie, the "coming soon" search engine, is anything but vaporware with a nice overlay of PR buzz. I just read an article that had the most concrete tidbit I have yet read - Dipsie's approach for indexing content from the dynamic pages in the "deep web".
|In particular, crawlers are stymied by dynamic Web pages, which are customized as users choose various options, such as car color at Cars.com. |
To counter that, Chicago-based Dipsie Inc. is developing software that promises to fill out Cars.com's simple online forms, which are based on multiple choice, though not the complex ones for the government's patent and trademark databases, which require typing in keywords. A public test version is expected by summer.
Mercury News [mercurynews.com]
Unfortunately that is all that Dipsie appears to be at the moment. No content. No search technology. No product. Just press releases and PR puff pieces.
|It's a shame, cos from a web design point of view, it could have great Public Relations potential. |
Yeah, now I'm starting to think that Dipsie was a search engine which never launched. They wasted lots of money in answering interviews, doing press releases etc that they had $0 in venture left, because they forgot to save some for the hardware/software.
I e-mailed them saying "wheres the SEO service, bro?" and didn't get a reply.
Yep, its all a tactic. What kind of a search engine helps in Search Engine Optimizing for other search engines? and makes press releases? a one which likes fooling around with the media.
What's the point in fooling around with the media?
Usually, the media is a tool used to widely promote or launch. How can a company be so silly as to blow it's operating budget on press releases, then leave itself with nowhere to go?
I wonder why someone would wish to perform such a stunt. To what end? What have they gained exactly by ding this.......
Microsoft is probably the most famous vapourware search engine [webmasterworld.com] on the planet. ;) We've even got a forum for it here.
Consider this: how much valuable feedback, product suggestions, etc might they have gotten by getting so much attention? A fair number, perhaps...and we all know, it's not hard to throw together an SE with about 1-20 million urls indexed.
And if they had that many to show people...but claimed it was "10 billion" - well, for many 'joe publics' they'll believe the 10 billion figure.
Throwing a Search Engine together offering 20 million results, is a bad idea. How many people can give suggestions without anything to go on (no dipsie engine).
People want one thing from a new net business:
An Original idea. I know people are going to say that there is room for all the engines, but I doubt that for a few reasons that you can't ignore.
Everything about the General SE's is the same. Also why would say loyal Yahoo/Google/MSN users going to suddenly swap over to Dipsie?
What is so special about it's services, and will Yahoo & MSN not be able to compete.
The thing is that everybody is talking and making assumptions about dipsie, when we don't know anything about it. There's nothing to go on except a press release and a bit about this bot.
Dipsie may well suprise everybody, but I doubt it. I'm saying this because from what I've read so far it's just another search box probably using someone elses results.
So what? What's new about that? Nothing.
If I sell you a fishy that tastes good, and my mate down the road sells fish that not only tastes nice, but he adds a new seasoning, to make it taste better........
My point is it's pointless talking about this stuff, until we have more to go on - at the moment there's zilch, nothing, nada etc
An engine can't make all these promises to save the world when there is no engine to see. That's no good -people want to see something.
If this was a promise from a company like Yahoo or Google, saying that we are working on a new engine etc with a 10b search index, I'd probably beleive them.
You ask why? Because these people have kept their promises in the history of search and they have their engineers and VC's etc who waste money on good stuff (as in the "technical or the "important" stuff rather than un-true "press releases").
That's because they're a company who know the true difference between true reputation and false reputation.
But common, an "unkown" media company, which has its own SEO service wants us to beleive that they are working on a new search engine called "dipsie" which will "revolutionize" the world with its "10 billion" page index?
Yeah Jeremy but Microsoft is actively spidering and has a track record. I'd say that Microsoft is keeping its powder dry and waiting to see how the Google flotation goes.
|Microsoft is probably the most famous vapourware search engine on the planet. ;) We've even got a forum for it here. |
Up to one million is easy. It could be done with a souped up desktop PC. When it gets beyond that, it becomes a more complex problem. You have to factor in spidering, rubbish removal, policing the index etc. Handling large datasets is something that a lot of programmers read about in college but it is a lot more complex than it first seems.
|and we all know, it's not hard to throw together an SE with about 1-20 million urls indexed. |
The big mistake that I've seen previous Irish search engine operators make is the one where they index everything without paying any attention to the quality of the index. I called them 'previous' because they don't exist any more. Admittedly Ireland is a small country and the number of active .ie websites is below 25K. Even the UK is not that much of a problem for building a quick search engine - I reckon that I could put an SE together with about 1.2 Million UK websites in about 72 hours (a lot of my work over the last few years has involved correlation of domains to countries (at a far more advanced level than a certain Indian site)). But that brings up the more interesting problem of site acquisition.
Google et al's approach to site acquisition is very like that old joke about an infinite number of monkeys at an infinite number of typewriters trying to replicate the works of Shakespeare. The standard approach is to start crawlers following all links on pages and then trying to sort the mess out. A refinement of this is to crawl the IP space of a country looking for websites. These approaches are incredibly wasteful in bandwidth and processing. The big, well financed, SEs can throw piles of money at the problem and come to something that approaches a usable SE. But for a small SE the possibilities are somewhat limited because they don't have the resources, bandwidth and finance to match the bigger players. The alternative is to go niche or to try and buy in data. And that holds its own Catch 22 - where is the SE going to get the funds to pay for this purchase? And more importantly would the revenue generated by the SE pay for the operation of the SE and the purchase of the data?
Some of the search results on Dipsie were in French. But interestingly they described Wiener as 'un programmeur autodidacte' - a self-taught programmer. This fact seemed to be missing from all the English language searches that I've seen. Most of the teams behind SEs have a good mix of business and techie people. But the important thing about the techies is that most of them have strong backgrounds in mathematics or technology. Most programmers could be considered as self taught as learning new languages and techniques is an on-going process for any programmer. This is why the reference to it sounds so strange. It would scare the crap out of venture capitalists because those guys are interested in Intellectual Property (patents etc) - look at the number of patents that Google, Yahoo, IBM, Microsoft are granted every year.
Maybe Dipsie is the next Google. Maybe it has an index of 10 billion webpages. Though it might just be vapourware. Even if it exists as a search engine, it has one fatal flaw. There is a sea change happening in the SE business. It is moving away from the macro-search engines such as Google to micro-search engines.
Macro-search engines like Google are great - they can find the most obscure character from some Star Trek episode that you saw once. But they cannot find the number of your nearest pizza delivery place. This is where micro-search engines come in. These are highly localised hybrid engine/directories that provide local, actionable information and data. Google and Yahoo with their mapping approaches are on the right road but they have not got it right yet. People want a limited number of correct and relevant results. Dipsie's anywhere in two clicks is just a marketing slogan. Without proper categorisation and correlation, 10 billion webpages is just a load of useless data. Even giving Dipsie the benefit of a lot of doubt, I think it is tied to last year's search problems.
Bravo - Exactly right.
People demand good results, after all who spends the time wading through even 10 to 20 pages of SE info?
It all takes time, to do the search, then work through the results so you find roughly what your seeking, and then sift through it again to narrow and confirm the searches so an approach to companies can be started.
It's got to take at least 30-40 minutes to finally get what your looking for. Call it an hour.
I like an easy engine, a nice engine. Technology is very flash and nice etc, but I reckon it can cause problems and confuse sometimes too.
Is Dipsie going to be very high tech or nice and basic to use?
Actually it is akin to the old medieval argument about how many angels can dance on the head of a pin. Since Dipsie appears to be all vapourware, your guess is as good as mine. (I was going to use the "That would be an ecumenical question." quote from the Father Ted tv series. :) )
|Is Dipsie going to be very high tech or nice and basic to use? |
| This 39 message thread spans 2 pages: < < 39 ( 1  ) |