« May 2009 | Main | July 2009 »
June 30, 2009 in Social search engines | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
Those of you who follow me on Twitter will know that I (along with Karen Blakeman and 10 other worthies) were invited to spend an evening with the UK Microsoft people to talk about Bing. You can follow up on the meeting by searching Twitter for #meetbing or just clicking the link. I've now had a chance to settle down again, and this is my take on the evening.
The Microsoft people are really nice; and give them lots of good karma for arranging the evening, and being prepared to listen to people. This isn't the first time that they've done this - back when they were launching their first iteration of a search engine I was one of the 'search champs' they invited across to Redmond to spend a couple of days looking at the offering. This was a rather smaller gathering and was supposed to be 2 hours instead. It's really excellent to see a search engine company/resource being that open, and it's a lot more than most of the others have been. They're also a very patience, good humoured and resiliant bunch - if I'd been them I'd have thrown me out in the first ten minutes I think. All told therefore I liked *the people* a lot. If you're actually from Microsoft though I suggest you stop reading here because this is about as good as it gets for you.
We were supposed to have a 20 minute ppt overview, break and demonstration. That went out of the window very quickly - 90 minutes in we were still on the 3rd slide. I don't think that they had anticipated the interest or number of questions that we all wanted to ask, and we quickly left the plan behind and just went into a Q&A session. I'd like to be able to point you to the slides, but Microsoft don't appear to be doing the Web 2.0 thing and they've not put their slides online anywhere, which is disappointing. Most of the statistics they used they had collected themselves via their own toolbar, which to be honest isn't any use to anyone, due to the intrinsic bias contained in that.
The key points that came out of the evening for me were:
Microsoft know that they're behind the game and are playing catchup. There was a lot of talk of a '10 year plan' which worried me; the impression that I got was that while they were in search for the long haul it really is the *long* element that they're looking at.
They want to be different from Google, but not too different, because that might scare people away. They apparently considered more graphical user interfaces, but decided not to take the risk. It seems that they're going to be happy to just raise the awareness of the product at the moment, and hope to leech a few users away from Google. Quite frankly I don't really see how this is going to work - if you like Google, that's what you're going to use. If you don't like Google you're going to be looking for a real alternative, and a Google clone with a few bells and whistles isn't going to cut it.
Their key market is the home user, rather than business. They're emphasising the ability to compare prices, doing searches for flights and so on, hence the vertical markets they're looking at. They also want to make a bit of money as well, which is reasonable.
I expressed disappointment that they're not being more radical and innovative - I seem to recall using the term 'woeful' to describe some of the things that they're doing, if for no other reason than almost every single one of them is already being done. This comes back to the 'being cautious' thing, which for me just screams out that as a company they're running scared and simply don't know how to innovate in search, and are just playing catchup.
I asked about the Microsoft censorship of UK results (the Martin Luther King search that I've previously blogged about) and they had no answer, but I'm promised one. They also didn't have much to say when I asked them about the blanket ban on searching for 'sex' in the Indian version. I also did some more searches at the same time 'penis' is a banned term, but 'vagina' isn't - so work that out! Also, as Karen pointed out, the video search option isn't censored. Now, in and of itself I don't really care that much about what Microsoft doing in the Indian market, and it's the choice that country to decide what they allow to be searched I guess, and I'm not really going to have too much of a go at Microsoft for bowing to local pressure. However, what does worry me is that they've just 'banned' the use of certain words in returning results. Just about every other search engine is capable of working out what's a pornographic site and what isn't, but clearly Microsoft has issues here. Coupled with the inconsistencies makes me doubt the ability of their team to pull together something that works properly. They also simply do not appear to have considered that the mouseover video starts playing makes the engine a porn lovers paradise, and I gently broke it to them that as a result Bing has already been banned in at least one educational grid for learning, and probably more. Yes, I know that they've added in the ability to block the specific domain that provides the porn, but do they really expect the average parent to understand what to do? I also expect that most IT departments will find it just as quick and easy to block Bing, never mind any specific subdomain. They just don't seem to have thought this through at all.
Karen, myself and several others expressed the concern that the functionality wasn't impressive. They've reduced functionality from Live (where's the link: command!) and they used to have a nice little slider option in MSN Search. That was at least innovative and interesting, but it's obviously too innovative since it's been dropped. We also said that we weren't happy that the advanced search option wasn't on the opening screen, and that a lot of functionality was command line driven, and difficult to find. This all got scribbled down as useful feedback.
Good news for the UK market - they seem to be going to roll stuff out to us next. We never did actually find out why we were starting out with less functionality that the US version, but there you go. I also understand that we're going to be getting UK images as background wallpaper but to be honest, that is almost literally papering over the cracks.
We did end up with a brief demonstration, but unfortunately that was of features that are currently only available in the US version, so the suggested 'good searches' of San Francisco and flight ticketing really was wholly inappropriate. Either don't show these, choose UK examples or wait until the functionality is rolled out here.
I think the Microsoft folks should have done more work on the group they chose. Clearly we all use search engines and were already familiar with basic search concepts, so they could have started at a higher level. 2 hours simply wasn't enough time, and I don't think the demonstration model they chose was appropriate. Microsoft - if you're going to do it, do it properly! One annoying sad little niggle, which I'm almost loathed to bring up - with all the travelling that was a good 7 hours out of my life, and I know that others came further. I'm happy to do that, because it's interesting, but ultimately it's in Microsoft's best interests, and our suggestions are (hopefully) going to help them improve their product. An offer to pay for travel expenses would have been appreciated, and the offer of Bing pens, notepads, cheap canvas bags and a 'Bing pong' table game was quite frankly insulting. Just another example that they're not thinking things through!
My overall impression is, as D Stuart said "On train home from meeting Bing...thoroughly nice chaps, not a chance in hell of beating Google" and I'd echo Karen's tweet "Would love them to have a fighting chance against Google but can't see it happening at present." I'll be really interested to see if anything comes of the meeting, but I have my doubts.
[Edited to add: I've now had an email with a request for expense details.]
June 30, 2009 | Permalink | Comments (3) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
Fefoo is a neat little multi search engine that provides access to over 250 different search engines. Straightforward interface as you can see below:
In most cases I found the choice of engines reasonable, although there were a few surprises - no Exalead in the websearch selection for example. When you run a search Fefoo loads the appropriate search engine into the main element of the screen, leaving room at the top for little ribbon search element:
Nice and easy to use, with a few command line switches for specific engines or types of search.
June 28, 2009 in Meta/multi search engines | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
June 27, 2009 in Google specific | Permalink | Comments (1) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
I just discovered BookSeer which is a discovery engine of sorts.
Painfully easy to use - just type in title and author and it will come up with suggestions for what to move onto next. It pulls content from Amazon and LibraryThing (though LibraryThing had thrown a moody when I was playing with this, so I wasn't getting anything back from there).
It seemed to work very well, giving about 10 recommendations from Amazon (and I presume a similar number for LibraryThing). A search for 'The Wasp Factory' by Iain Banks suggested 'The Crow Road' by the same author and also 'The Bell Jar' by Sylvia Plath. 'Here comes everybody' by Clay Shirky led to the suggestion 'Wikinomics' by Don Tapscott. 'How to use web 2.0 in your library' by, errr, me, lead to 'Library 2.0 and Beyond: Innovative Technologies and Tomorrow's User' by Nancy Courtney.
Nice resource - simple, easy and quick to use. Great to have easy to hand at the public library enquiry desk when the question is asked. However, if you don't like this you could try:
What Should I read next?
Literature Map
Debbiesidea
Booklamp
June 23, 2009 in Utilities | Permalink | Comments (2) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
Imagine that you were in a situation where you had to collate information on the names, descriptions of, dates of birth for, and birth state for American Civil War generals. Or similar historical data for English castles. Or perhaps data on African countries, with descriptions, capital, currency and size. It's not actually that difficult to do it, and you can pull the data from various sources with ease except that it's probably going to be a little bit long winded and could keep you busy for a while.
Now imagine if Google could do it all for you = create a matrix, compile the content and just pop all that information directly into a spreadsheet for you. That's the idea behind Google Squared. Simply pop to the site and type in your query 'African countries' or 'Civil War generals' and Big G will populate a spreadsheet for you. It'll do it automatically without further intervention on your part and will provide access to things such as images, currency, dates of birth and so on and displays what it's created.
Marvellous! If you can't see the data that you're after then you can add more columns, and Google will go out and find the new data and pop it into the grid for you.
That's the theory. In practice, once the excited searcher (that's you) deviates from the carefully chosen examples Google offers things fall to pieces very quickly. In fact, you can see this already in the screenshot above. The capital of Tunisia is apparently +1hr, begins last Sunday in March... well yes, quite.
This happens quite frequently - either no data is returned or very often it's the wrong data. My Civil War generals search started returning information on battles, it was unable to work out the full name of generals, even though they were highlighted in the description column. You can do a mouseover to see the grid which then reports the origin of the content, and Google is pulling this stuff from all over. In a sense this is good, but when one biography comes from Wikipedia and another from Amazon you have to worry about the comparability of the data. This in turn means you have to go and check everything individually. A square for Civil War battles turns up Abraham Lincoln as the first element, while the result of the battle of Balls Bluff is 'A small'.
You can overcome this to a certain extent by creating your own square and populating it with whatever appeals to you, and add in any other content as appropriate. This is actually more difficult than it sounds. English castles perhaps? Or England Castle. Or (england OR english) (castles or fortification) -"hill fort" instead? You quickly get the drift, I'm sure. Keeping it simple I get the name of a castle, a good image, a nice description and then it all falls to pieces. 'Country' is the next suggested information column, which is pretty redundant. Location is blank in all boxes, and List of places as the final column is meaningless. However, let's get rid of those and do something more sensible. County works quite well, but 'telephone' fails, giving me American numbers or nothing, as well as accurate results. 'Phone' works better, but gives different results to the 'telephone' option. Trying the combination of telephone OR phone results in zero results. How do I discover when building started for the castles? 'Begun' gives nothing, and 'started' is the same. 'Date' does give me dates in some instances, but they're all referring to other things. 'Built' however is much better and does give me something to work with. I also had better luck with entrance fees, since my first value 'ticket' wasn't too bad and only failed on 2 out of 7 castles.
This is the problem with Google squares - you have to work out appropriate terms, and yes I agree, you have to do this when you're searching, and if you get it right that's great, but if you get it wrong you have to choose another value and add it in or start again, and if you get two different values (as I did with the phone/telephone example) you've still got to go off and check. Moreover if you leave it up to Google to choose values there's no consistency or obvious way of working out what it's done.
You can save squares to your own Google account, but I didn't see a way of exporting the data anywhere else, which is just madness. If I was doing this 'for real' I can see that I'd need another tab opened or flick between browser and a spreadsheet package to pop the details in.
It's a lovely idea, but it simply does not work as it should. Even in beta it doesn't work, and far from reducing time spent constructing a matrix it just adds to it. I really wish Google had spent longer developing this before releasing it to disappointment, but then that's Google all over isn't it - why worry about quality when you can make a publicity splash? I do hope they continue to develop this application because with a little more work it could be genius. As it is, we're looking at another idea frittered away against a wall. Tragic.
June 20, 2009 | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
June 19, 2009 in Search engines | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
Clay Shirky is always good value to listen to. Here is a recent TED video where he's talking about the ways in which communication is changing, using recent examples from current events. (Thanks to Dr Alan Cann for alerting me to it!)
June 19, 2009 | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
June 16, 2009 in Google specific | Permalink | Comments (0) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
June 16, 2009 in Meta/multi search engines | Permalink | Comments (1) | TrackBack (0)
Reblog
(0)
| | Digg This
| Save to del.icio.us
|
| |
Recent Comments