In one sense, Danny Sullivan publishing a story about Google accusing Bing of copying their search results on the eve of the Future of Search event that Bing is co-sponsoring that derailed the entire discussion on search quality and became a running joke throughout the full event - without giving Bing time to give a full response - is perfectly innocent. Google pitched the story to him the previous week but he didn't have time to get a briefing earlier so he ended up writing just before the event.
When you have a good story that you want to tell, from a good source, you want to publish. You might get scooped, you want credit for breaking news, there's even an element of excitement in getting the story and putting it straight out. There's a reason it's called breaking news.
But that's not why Sullivan says he posted when he did. He seems to be saying he posted straight away because Google wanted him to. "Yes, they wanted the news to be out before the Bing event happened — an event that Google is participating in. They felt it was important for the overall discussion about search quality." They didn't feel it was important for the overall discussion right after they did their experiment (which Shum calls a form of clickfraud; the gloves are off in this argument), which gave them results on Bing at the end of December; Google didn't offer Sullivan a story on what they called search relevancy until January 13th. And they didn't approach Microsoft to ask what was going on; they went straight for the publicity, right at the time of the Bing-sponsored conference.
And while I agree with Sullivan when he says his site isn't there to do PR for Google or to do PR for Bing, I think he just did do PR for Google by letting the source dictate the timing of the story. And as the executive editor of the New York Times put it just last week when talking about Julian Assange: "The relationship with sources is straightforward: you don’t necessarily endorse their agenda, echo their rhetoric, take anything they say at face value, applaud their methods or, most important, allow them to shape or censor your journalism. Your obligation, as an independent news organization, is to verify the material, to supply context, to exercise responsible judgment about what to publish and what not to publish and to make sense of it."
(And no, I don't think it's the same as taking information under an NDA - NDAs say I can't write about the topic before a certain date but don't force me to publish as soon as I can , although they do encourage it).
By serving the agenda of the source, you miss some of what the story could be. I've lambasted Microsoft for not joining the conversation by giving comment on some stories, but you have to be realistic. When you ask for comment on a big, controversial story on a day when some of the senior people in the team you're asking for comment are preparing for a big event and it's not something they already have a view on, you're probably not going to get a very detailed reply. I've asked fairly complex technical questions that have taken weeks working with the developer to get an answer on, because they literally didn't know what was happening. Unless this was a Sekrit Bing Plot All Along*, (in which case I'd expect there to be been a plausible denial on file ready to deploy), someone at Bing is going to have to look at the claims, find out what's happening inside the engine, find out why it's happening and decide what the policy is on that and what to say about it. And they have to work around whatever they had scheduled for that day, which just happens to be the day before the event when they're likely to be pretty busy preparing from.
That's pretty much what Sullivan seems to be saying happened. "Bing was allowed to have as much input as they wanted. I contacted them early on Monday morning, when I started working on this. I received a reply around 3pm Pacific, which I included in this article. That’s all they wanted to provide... I went out of my way to follow up even though I’d already been given a statement... [the spokesperson] said it was likely they’d have more to say when I see them after the event today."
If Sullivan had waited to post, with Bing's flat denial (kudos to Mary Jo Foley for getting an unambiguous quote to go with the explanation) and the extra details from Harry Shum's post, the article might have been less punchy (it wouldn't have had the line 'Bing doesn't deny this' in the first paragraph) but it would have been better journalism.
Would it have been a letdown after the Farsight event? Hard to say because without the article the discussion at the event would have been entirely different; the stories might have been about what search engines can do in the future rather than whether they're playing fair right now. I think a 'did they cheat' article would have had as much impact after the event - but it wouldn't have given Google the same level of exposure for the accusations and it wouldn't have turned an event that Bing had sponsored as an open forum for many different companies in the search arena into an open sparring match. This is pretty much journalism as spectator sport, and I like to think our profession is better than this.
* On balance, I don't think Bing was deliberately taking search results from Google; I've met enough Bing folks to believe that they they're smart enough to know that if they did that, they would get found out sooner or later and they'd face exactly this kind of backlash. I think they're getting those results from Google exactly the way they say they are; by mining user data, and no-one ever thought to go in and censor the Google search strings in there. Maybe they should have. (Maybe Google should filter out a lot more of the content spam sites that make money from Google ads before they get complaints about them rather than waiting until the results are so polluted that people complain; everyone is making a judgement call on these issues). I agree with Shum; Google uses user data (like the content of your email in Gmail) to apply ads, the Google toolbar has a long EULA just like the Bing bar and the IE 8 suggested sites feature - and as I've said a lot recently, when the service is free the users are part of the product being sold. Bing is trying to get better search results by looking at the pages real people visit - I wonder if the reason that only 7-9 out of 100 honeypot terms made it into Bing is something to do with the behaviour of the Googlers baiting the sting? - and real people use both engines. Maybe what the 'sting' actually reveals is that search results have a massive influence on where we go on the Web, and that is the point at which the search engines needs to take responsibility for the results they provide.
I know tying is a key part of monopoly abuse, but I think reinforcing is more important. Take the deal where Google translates the patents for the European Patent Office; it gets paid, it gets to index the data to extract knowledge it can use in the search algorithm and it gets parallel documents it can check and then use to improve its translation algorithm (I'm just assuming the translations will be checked because even in a closed domain, unchecked machine translations of patents sound like a lawsuit waiting to happen).
It's not a crime just to be big; the issue is what you do with your size and whether you do things that unfairly keep others out of a market.
Everything Google does in all areas of its business (apart from possibly Google Apps in enterprise and government, and it may apply there too) is designed to either put Google ads in front of more searchers on more platforms or to give the Google search engine more training data in order to improve the search engine that puts ads in front of searchers.
Gmail, Picasa, translation, the Google Maps sensor probes you have to take to put Android on the phone you sell – they all get more data for Google to crunch. Their business model is transforming the information of the world into a source of targeted ads. By having a bigger index, Google gets to have a bigger training set for its engine. And how much it promotes its other services that funnel training data to the engine has to matter as a question of reinforcement; the circular nature of what Google is doing doesn't work in isolation.
You can be too real time: Google vs Bing
Feb. 23rd, 2010 08:59 pmTake 'mike siwek lawyer mi'. Steven Levy got an unusual level of access at Google, and covers the different stages of what Google has used in its search algorithm over at Wired. He does it rather uncritically; it's a piece about Google not an analysis of the state of search and he skips technical analysis of Bing in favour of colourful metaphors. One of the the things he quotes is Google's claim to do better on a specific search as an example of how it's better on names than Bing; search for those four words and Google thinks you want a lawyer in Michigan and puts him at the top of the list, Bing, says Levy, doesn't get him for several pages. But search is a work in progress; Bing now has several ways to find the lawyer in the first few results, including directories of lawyers. What does Google actually find tonight? A lot of references to Levy's article, to tweets about the article - and no link on the first page of results that brings up the man himself.
No single result is a good test of search; I keep both Bing and Google as search providers and maybe one time in ten Bing doesn't give me what I want and I repeat the search on Google. I'm sure in a few days, Mr Siwek will bubble back up on Google. But Bing's results aren't broken by those 'more important because they're recent' stories the way Google's are; there are links to Levy's piece but it doesn't assume that's all you want. I suspect that's a different emphasis in the search tuning and usually, that's more useful. There's a whole range of what makes things interesting and novelty isn't the only measure.
Early adopter: check. (Well, I think it's more fostering, given that I don't keep most of the tech I try out; journalists - fostering today the tech you'll want to adopt tomorrow?)
Shop at Walmart: sometimes. Their clothing line has some nice stuff sometimes, but I'm more of a Trader Joes/Whole Foods gal.
Am I more a Google searcher? Target and Amazon? Well, I bought a dinner set at Target and dragged it back from Cincinnati... certainly not a Yahoo users and my views on AT&T are, well, frank... (capex down in almost direct proportion to increasing revenue? file under shortsighted)
AdAge has some fascinting demographics from a study by Wunderman ,BrandAsset Consulting, Zaaz and Compete: I'm not sure about the search engine users but they are *spot on* about the AOL demographic; that chip on the shoulder, 'I'm comfortable here and I'm staying but shouldn't there be more' attitude - they have always been the core AOL audience. They were often obscured by the transient wave of people using AOL as training wheels to get online, check out the walled garden and hike out into the wild Internet. Part of me is thinking I'd never want to build a brand on that demographic, but if you can pitch them, keep them and sell to them - why not? The problem for AOL is that it's a market that isn't sexy, doesn't look good in headlines and somewhere along the line tech and online services have become all about what looks good in the headlines...
What Your Choice of Search Engine Says About You
"What does your search engine say about you? Well, if it's Bing, you're probably an early adopter, but you also visit, shop and ultimately make purchases from Walmart more than other search-engine users. Google searchers, on the other hand, are partial to Target and Amazon, and Yahoo searchers have a strong preference for wireless service from AT&T and Sprint.
Google users are more likely to book a flight online at JetBlue or make a reservation on Hotwire. They are also more likely to do research on a Lexus, while Bing users tend more toward Toyota.
For instance, AOL customers feel less intellectual than their peers, are 55 and older, spend their money more responsibly, want to blend in to the crowd, feel like they've gotten a raw deal out of life, expect less from their future and, believe it or not, still use dial-up modems. Bing users are middle-aged, highly educated tech-savvy individuals who consider themselves to be average and spend more than 10 hours a week online.
Googlers tend to be the average internet Joe, according to the study. The search leader's loyalists are conventional people yet open to trying new things, believe in following rules and don't consider themselves any smarter or less intelligent than the person next to them. Yahoo users tend to be 55-plus, reserved and a less-independent group with little faith in imagination. They feel they have little control over their future and are skeptical and cautious of new or untried ideas."
Google and privacy again
Jan. 27th, 2010 03:20 pmMaybe I should have a privacy_rant tag to save time...
The new version that Bing asked me to download this evening sports a new (grey) interface with Bing styling, instead of the colourful - and nicely obvious - icons. Movies gets a big ad at the bottom of the screen; just like the old icon for finding movie times but somehow more important than anything else Bing can do (judging by the fact that it's twice the size). News pushes its way in, because obviously I want to search for that on the move more than anything else (not). And the 'always a day out of date but still incredibly useful' service comparing gas prices at local garages? (i saw gas not petrol, because it only worked in the US). Gone. It's not even hiding in the transportation section, which has no category for petrol stations, gas stations or garages of any kind (Microsoft seems to be suggesting you park the car and take a ferry to reduce global warming); it's just gone. As is the option to tell Bing when I'm in the US and want to get US services and results and when I'm in the UK and don't. So that's about half as useful for a start. That's a first; updating an app and making it less useful. Bing is all about taking data and turning it into useful services, so why take them away from the phone where you need them most?
This is as bad as Microsoft unveiling its other flagship Windows Mobile app, the beta of Office 2010 and giving it the sum total of maybe 5 new features (and a business only Sharepoint app). Dear Microsoft: it appears to have escaped your notice that the smartphone marketplace has a great many consumer users in, and they're increasingly picking other platforms? WinMo has 9% market share, 3% share of apps and Web pages with ads on served through the AdMob network and a glorious future behind it. Or more briefly...
Dear Windows phone team. You're doing it wrong.
Google news by author
Jul. 1st, 2009 03:43 pmFor me it finds recent IT Pro posts, and a story on The H in the current results; in the archives it says it goes back to 1998, though the first story is from 2007 and the second, though marked as 2008 is actually from 2004 (the disclaimer at the bottom says 'Dates associated with search results are estimated and are determined automatically by a computer program'). 1998 marks my time at AOL, when I was quoted in press releases as well as writing about them; 2002/3 has me in Computing, reviewing on Amazon UK and writing for the Taipei Times (it's actually a reprint from the Guardian) about the first flood of spyware. By 2006/2007 Google has a lot more of my writing - for some reason November 2007 was a very quiet month. For 2008/2009 it includes my annual company report (who thought that Experian would be a news publisher rather than a primary source) and it doesn't find any of my articles on TechRadar or Tom's Guide. I'd say maybe 60% of the articles that I write that go online are being picked up as written by me, so I'd give it 6/10.
It did find a few stories that turn out to be about me, at least peripherally; http://www.theinquirer.net/inquirer/news/1032699/british-hacks-urge-intel-to-raise-chip-prices made me chuckle...
My friend Peter asked me over lunch on Sunday how Simon and I advertise our blogs, because he's starting to blog more and wondered how to attract readers. Peter gets a page on Wikipedia and he's interesting to read (go argue with his latest post about whether Linus supports DRM), but he noticed that the post he wrote that got picked up on Boing Boing got a lot more comments than other, very similar posts (indeed, when you search for him on Google, you get, in order, the Wikipedia link, a ZDNet blog, his Linked In Profile and a link to just that post).
'I don't bother with SEO,' I said; 'I'm mostly talking to people who are already listening.' Which probably indicates a certain lack of ambition. SE-what, he replied? Well, now I can point him at the SEO-by numbers explanations in this blog on IT Pro, which sums it up as lots of links, phrases people will search for and keep the sentences short so the spider can understand them. There's a company raising venture capital on the basis of promoting link journalism, where you annotate links to other online articles to synthesise the evidence and put your own view in context (kind of an accessible annotated bibliography). And the blog also links to a superbly subtle column by Charlie Brooker for The Guardian that criticises the practice of shoehorning in irrelevant mentions to Britney Spears and Angelina (and of course, in the process, includes them often enough to squeeze out the very Google juice he's decrying).
The comments seem to rather miss this parodic point and criticise the column for being too ivory tower. 'No one is stopping you from having a beautifully crafted article that no one can find,' says one but then claims we need to optimise the superior content to make it stand out from the masses. Quite how pumping it up with the same search steroids will do this, I'm not sure. I know SEO works well enough for people to pay for it, but much of it strikes me as the Emperor's New Clothes of the Web. If everyone with a site on topic X optimises it with the right keywords, how will any one of them stand out? And isn't it actually about making it easier for us to stop assessing the value of sources ourselves, to stop seeking out good writers and just listen to what shouts the loudest and flashes the brightest?
Very little writing is pure, self-indulgent art - except, ironically, blogs. Anything that's written for publication has to make a point, fit a format, reach a reader (yes, alliteration and metaphor help). The length of fiction - flash, short, novella, novel, series - dictates both the depth of plot and character and the structure of scene and climax. News stories have to read from the top and cut from the bottom so you can fit what matters on the page at the last minute - call it a dying art, but more people in the UK read UK newspapers on paper than online. Sidebars and other page furniture break up a magazine article; they're called page entry points because they can get you to start reading, but they can also give you a break from following the thrust of the article and let you absorb it before you return to reading. Any piece needs to speak to its audience; if I pick up a romance, I'm not after the gritty, indulgent gore of true crime. So you can't say 'it's art, I shouldn't have to care about the demands of commerce or prostitute my pure writing to structure'. But much SEO structure is banal and reductive and lazy and lowest common denominator in a way that other writing structure strictures aren't.
What about good SEO structure? Probably, I've seen it and not known I was seeing it. Maybe, like any other writing craft, ars est celare artem.
Mobile search, mobile work
Jul. 24th, 2007 07:11 pmWhen you're on the move, do you want to search the Web the way you would on a PC, or rather look for what's around you? Sometimes you'll want to look up a Web page and read it, but often you want to know more where a movie is playing rather than who was in it, where to get good sushi rather than how to make it, and how long it will take to get to the theater after you've eaten. Read the rest of Simplifying Mobile Search...
Need a bigger screen? Thin and light or mobile workstation, basic budget or high-powered business features, Macs or tablet PCs; today we’re going to tell you how to choose the right notebook for whatever you need. We’re going to go through business, general-use, budget, gaming, ultra-portable, tablet and Mac laptops to show you what to look for and offer some suggestions. Pick the Perfect PC for You...
Make Vista search mapped drives
Feb. 27th, 2007 07:16 pm![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
Microsft's official stance has been that searching remote drives slows things down too much; they have to fix that when Windows Home Server comes out. There's still no option to snooze or restart indexing in Vista the way you can in XP: a little too nanny-knows-best alas.
Finding my articles on the FT
Feb. 27th, 2007 03:56 pmhttp://search.ft.com/search?queryText=mary+branscombe&aje=true&dse=&dsz=
In other news I'm finally out of bed but not really back up to speed - still coughing a lot and not really running on all cylinders brainwise; reading and watching the brainrotter are about my limit.
My Financial Times articles
Jun. 8th, 2006 01:37 am![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
zvents is very pale-blue-and-orange-with-white-space Web 2.0, similar to Eventful: I'm seeing this look a lot at the moment. Unlike , it only covers the Bay area. You can search by events, venues, tags, groups or people, and when you get the results you can switch the list to a map view or a calendar view. With a lot of results the map and main calendar view show you the number of results rather than the individual events, but you can get a 1 day, 3 day, 7 day and 30 day view as well. Here it's the guided tours that dominate rather than the bookstore events, but when you get down to individual events they're very well presented with maps, times, repeat events, similar events, other events at the same venue... More useful details than Eventful.
AllConferences has a hierarchical drill-down of categories and an advanced search, though you can only search by one condition and picking March 2006 without a date produces events from June 2004 as well. There are conferences going back to 2001 and those are what you see when you search by City; the general search box does a better job. Look here for commercial and academic conferences.
These sites tend to be better for finding a specific event on a specific day than browsing through the possibilities for a longer period of time. For that, I want to be able to start with a large pool of results and filter them. The best filtered view of search results I can think of - and it has deficiencies still - is the hotel map view in Expedia; you can zoom in on the map to refine the list of hotels, or remove hotels from the list to clear them from the map view. I'd like the same for events; let me zoom in to an area, or a category of events, or a smaller date range, or to a time range across several days (what's on every evening next week?). Let me remove all the sports events and everything that's recurring rather than a one-off and trim down from any day in March to just these 9 days. It's all about underlying hierarchies of logical units: know that a week is a logical unit of a month, know that Kirkland is within the greater Seattle area. Some of this you can do with a folksonomy, but a categorised hierarchy is going to help for geography, discrete units (today/tomorrow/this week/next week/this month/next month/this year) and distinguishing between broad tags (music) and specific tags (baroque). Organic tagging can define a problem space, but it doesn't structure it well.
Do we have these kind of detailed schemas for describing not just the obvious properties of events (date, time, venue, organiser etc) but also the range of values so we can build the filters?