How perceptions of Being Bigger may not always play out in reality.

First Mover Advantage. A popular adage in this era of perennial web startups. Equally as strong has been the thought that established players – the big fish – can simply move in on the small fry’s niche patch and bring it to the mass market – the Free Rider effect.

Groupon is arguably a case in point. Out of nowhere the young gun from Chicago has been described as the world’s fastest growing company and a $6bn gamble, only for such heavyweights as Google and Facebook to begin to muscle in on it’s coupon territory, sensing their already huge ecosystems will at least make the new guy irrelevant.

This is the textbook case. What happens when the innovator is somewhat higher up the foodchain? We may be about to find out.

Google has accused Microsoft’s search engine Bing of ‘monitoring’ their results for search terms and seeking to mimic them on it’s own site. While kudos go to the Googleplex for gaming their own search engine to find proof, it’s interesting that Microsoft feel that Free Riding on an innovator the size of Google can actually work.

Think about it, what hold has Redmond got on your computer these days?

Apart from the difficult to ignore growing band of Apple eaters, personally I can’t remember the last time I used Office or IE, both systems previously heavily relied upon by Microsoft to cross sell/arm twist users into other services. Sure they’ve still got the OS. But then the legal battles have all but robbed them of the leverage this previously gave them.

Google on the other hand have a rapidly developing web services suite that is keenly integrated with their search – when you sign into Google, you get all their services without downloading one byte more.

Mimicking Google’s search will do Bing – and by extension Microsoft – absolutely no good whatsoever, because their ecosystem is no longer the trump card it was.

Indeed if informed opinion is anything to go by, it should be attempting to differentiate and not play copycat.

Google has done two ‘real-time’ things lately, one good one not so good: Real Time web indexing and real time web search.

With ‘er, hang-on a minute…‘ moments now surfacing in the public domain I find the contrast between the two to be especially important. Google in their traditional engineer style expound the benefits of both in shaving seconds of search: ’11 user hours saved globally each second’; ’50% faster indexing rate of content’; figures that prove the mantra – machines search better than humans.

Machines definitely do the donkey work better than humans. Indexing is a dumb process easily solvable by machine and has been for decades. The migration from batch processing to incremental updating of the search index that Google Caffeine delivers is an essential improvement to real time search.

The Google Instant realtime GUI trick is not such a homerun. Instant brings up a full page of results updated character by character. In cases where the user searches over two or more words – in my experience the vast majority of search – context is vital. Rarely is that context clear until the entire phrase is typed in. This is why google instant, as fast as it undoubtedly is, rarely returns what you’re looking for until you complete your search term.

In any case, the Mind Machine Interface is a delicate thing and only as strong as the weakest link – the human. And it’s the human that has to comprehend this extra flow of data, most of it extraneous.

Google does not yet do the contextual understanding the user must accomplish to use Instant search successfully – and I wouldn’t like them to try, as that would likely involve personalisation based on past searches and as my browsing habits change over time I don’t want past results skewing things.

Incidently an ulterior motive for Google Instant can always be found on the web.

So in conclusion real time is only useful when the data can be transformed into a form easily processed as by the end user. If it cannot it instead serves to exacerbate the problem of information overload rather than lessening it.

The ideal real time UI has yet to be realised.

Naked Domains

Due to load-balancing based around CNAME records, Google won’t let you have a nice minimalist http://example.com url for your app. Just set up a 301 HTTP redirect in your .htaccess file to the de facto standard subdomain ‘www’.

Unlimited Apps

This problem was exasperated for a long while by the fact you couldn’t remove apps once created. Thankfully Google now have a delete button, but you still don’t need to use it. Simply register a new Google account and use that to gain another 10 apps. Before you do though, apply for a free mobile pay as you go SIM card that you can then use to receive the validation SMS necessary to activate the new account.

]]> http://aleatory.clientsideweb.net/2010/08/23/some-app-engine-hacks/feed/ 1 http://aleatory.clientsideweb.net/2010/07/12/google-its-search-for-a-social-graph/ http://aleatory.clientsideweb.net/2010/07/12/google-its-search-for-a-social-graph/#comments Mon, 12 Jul 2010 02:39:54 +0000 rutherford http://aleatory.clientsideweb.net/?p=296
image courtesy Tantek

Techcrunch lead on a $100m investment by Google in Zynga the social network gaming company. It’s the latest in a long line of Google failures in a vital area of capturing web traffic – or, “organising the world’s information” as Google diplomatically puts it.

Canny move or desperation?

When examined in detail for such a blazingly successful company as Google are their record in social really is shocking: a near-continuous stream of failure with little apparent success on the horizon anytime soon. Something is playing up in non-evil algorithm land.

Dead

• Dodgeball
• lively
• Jaiku

Not Dead

• Latitude
• Wave
• Buzz

All are relatively unsuccessful on either Google or their own niche terms.

All elements of organising a user’s network around their infrastructure and not someone else. Each one concentrating on a certain social metaphor and each one failing in their own anonymous way. What will be different about gaming? What will make users skip their existing social accounts and networks that already have these games? While they tend to be pretty limited in depth what will Google do that will be different?

Obviously I’m highly sceptical but when you look at the stakes it’s vital for Google that they find a social something that sticks. Especially when the CEO has always seen them as a media company first and foremost.

Facebook with their über-targeted ads has a technology and reach that is arguably more powerful than Google’s. It is a big threat to continued adwords growth, which despite all the moves against Microsoft & Apple is surely Google’s #1 priority.

If gaming really is the leverage to unseat Facebook then Zynga may turn out to be a great choice – the 3 year old company is already rumoured to be projecting $1bn in revenue for 2011, signalling the undoubted importance of this much derided gaming sub-industry.

It may be blood from a stone attempting to get Google to admit it’s worried about anything – least of all a single web site – But putting yourself in their shoes it’s difficult to see Facebook’s rise as anything less than the #1 threat to their existing revenue model.

App Engine is generally a new paradigm for webapp developers; replacing sessions with memcache and a schemaless datastore just two elements requiring new thinking for old problems. Unfortunately there are a few more hidden nuisances which have the potential to waste programming time relatively early on. Here’s four of my personal head-bangers:

1. the datastore doesn’t always store Properties
I’ve had trouble with it refusing to store arbitrary entity props unless I assign them in the entity constructor itself (these fields were optional btw). Just setting prop values after initialisation then put() on the ds didn’t write them.

2. fussy filter parsing

.filter("prop= ",propValue).fetch(1))

returns a NoneType error

.filter("prop=",propValue).fetch(1))

silently fails to find expected.

.filter("prop =",propValue).fetch(1))

Only the above will return the expected result.

3. Only possible to execute inequality filters on one property per query
This is a pain in the arse if you want to query whether an input date is between two dates stored in a particular entity – officially there was a workaround whereby the date range is stored in a ListProperty (instead of two fields of type DateProperty) and you do the normal check if input is more than the list (greater than at least one element in the list) and less than the list (less than at least one element in the list).

However the App Engine team has now changed the behaviour in the cloud whereby both the ‘>=’ and ‘<=’ filters are operated on each individual list element and not a lazy test over the whole series i.e. where the existence of two list elements that bounded the input date would have been sufficient the following query now only returns the entity if one of the ListProperty elements is an exact match for it:

WHERE date_range >= :1 AND date_range <= :1

Unfortunately this has not been removed from the dev_server datastore, hence it runs perfectly well locally.

4. And hopefully the 1000 record query limit is well-known by this point.

On a more general note, why is it newfangled tech doesn’t build on top of the old stuff? Re-use would get us to where we want to be a lot sooner. I bitched about this on Twitter at the time and I’ll repeat the message here too because it’s worth doing so frankly, I expect new stuff to do the same things old stuff does as well as any “hey that’s cool” new fandangomatrons it bolts on.

Wave’s another case in point. Not ‘email invented today’, far from it – it’s left so much cutting-edge crowd-sourced participatory stuff out (as well as how to do IM, namely the KISS principle) – that it actually feels like a retrograde step in many ways. Most in fact.

Bottom line – Google needs to make stuff better.

Google appears not to add pages that include the rel=”nofollow” meta attribute by default.

I submitted the url manually more than a few times in case google was failing to pickup the sitemap for some reason. None caused the post to appear in Google search results. Then I went a bit deeper and analysed what parts of the post were different that would cause this. The only structural change to my posts I made where the existence of the nofollow meta attribute on external links referenced in the article.

Although I had introduced this a few posts earlier (chiefly to prevent the mass of links present in my Friday linkdump from downsizing my pagerank), I guess Google give it a few posts before such content starts ringing alarm bells. And I know PageRank has become less relevant now compared with search page ranking according to specific keywords, but I like fiddling with these things.

So I removed the nofollow attributes and resubmitted to Google. The result was immediate – #1 post for the ‘google deadpool’ query on Google.com. So don’t use nofollow within your post content.

Dodgeball
This is what happens when the big G ventures outside of it’s tech stronghold: Tech invents something better and Google is left wondering why it got itself into some arcane mysticism it never really understood in the first place. Dodgeball, social networking on mobile before smartphone-based GPS services went nuclear – it even used SMS. Bless.

Google Answers
Two for the price of one here, as Google Q&A was a forerunner to Answers, and whose comedy value is far more interesting. Not today’s Q&A app that very few of you may be aware of, but an old web0.1 version whereby users emailed Googlers who responded for a fee, presumably with some answers. We never found out for sure though, as the service only lasted 24 hours. Nothing like fail fast, fail often eh Google?

Click-to-Call
More phone-based shenanighans here. Enter your number. Google calls and puts you through to an advertiser. Gave clickfraud a social networking element.

Google Browser Sync
Somebody must have been seriously running out of imagination for their 20% time. Users could upload Firefox cookies, bookmarks, history and transfer them between machinezzzzzz…

Google Web Accelerator
Although you can still get this on those software hoarding sites, it won’t work. Not a bad idea when the web ran on hamsters, but then Google held back 5 years and released it when there was no need for it. D’oh.

Lively
I get social networks moved online. I don’t get the obsession with reinventing every inane part of daily life. I don’t want to walk around a room online to meet people – it sucks and completely misses the point of well, not needing to walk around a room online. For this reason 2nd life will die out eventually too.

Walking around rooms != fun m’kay?

Jaiku
Twitter for the über Googler. I know it’s not dead, but ‘open sourcing’ something after pulling all your developers off it when you’re Google it is like throwing the chef off a destroyer class warship onto a life raft. In the middle of the Pacific.

Google Audio Ads
Yeah let’s do radio too. No. Next.

Google Notebook
Taking notes whilst browsing was handy. Especially handy for not a few people was the nifty firefox extension. Unfortunately though Google were developing a hundred thousand different wordsmith applications all at the same time. Redirect -> Google Docs || Bookmarks || Gmail Tasks.

Google Video
I have watched 4,653 videos on youtube. And I’m someone who has the flash furies. Nevertheless it pissed all over Google’s own in-house attempt at a DDOS attack on the world’s web browsers. No longer accepting uploads means you’re dead vidyo.

Mashup Editor
A cool idea, but a me-too one. And the original, Yahoo! Pipes, was (and more importantly Google still is) way cooler.

This could be a turning point on the web.Vic Gundotra, Google Engineering VP

On seeing a screenshot my first thought was ‘Facebook news feed’.

It’s got a similar albeit primitive system of inline comment right now. While it’s a nice to have, it’s not ‘killer’ and some of the Wave features (every character typed is spewed out in realtime as if participants were viewing the same terminal) will turn a lot of people off straight away. Indeed I’ve a few ideas of my own regarding ‘next generation’ communication, and they don’t involve greater intrusion. While the inline editing is a good thing I don’t regard it as a big enough win to justify moving onto a new platform, cloud or no cloud. What would be wrong with sticking this kind of functionality into Google Docs? We need smarter comms not more of them.

First quality flame post appears to be at Gigaom…

• Twitter  One of the hottest web properties today.  A resounding success thanks to FMA.  Just don’t mention the business plan.
• Friendster  Big in the Asia-Pacific region (I’m not sure if that’s the web equivalent of stating a musician is “big in Germany”) but has fallen behind FB & MySpace in Western online society.  Imo FB’s rep for being a college student’s thing got it viralling and the API sealed it’s dominance in the long term.
• iPhone App Store  Certainly got developers excited, but then wouldn’t the devs who like to be known as creatives already have iPhones in the first place?  The proliferation of announced app stores for the various mobile platforms since merely means the commoditisation of the idea and another wall around each silo.  First mover advantage will mean little in this regard.
• Amazon Elastic Compute Cloud  Amazon went at it full pelt – access to root, command line, the whole server OS.  A brave move.  Second out of the blocks was Google’s App Engine, offering a much more restricted/simplified Python-based service depending on how you look at things.  Crucially though it allows devs to play for free – that together with the “it’s Google” factor give it a much greater monthly pageview traffic than EC2′s homepage.  But it’s cpu cycles that measure the success here and EC2 already bears the hallmarks of an enterprise-ready app in a production environment with a standard SLA in place.  Too early to call then, but investors have been asking for more visibility on Amazon Web Services in general, so definitely a future battleground.

Has first mover advantage disappeared then from the web?  Although many breakthrough applications have been superceded by bigger and better alternatives with a far greater leverage, none are by any means rendered dead in the water because of it.  And neither are the fast seconds always some massive corporation throwing money at the latest buzzwords; Zooomr established itself by providing a slightly more interactive interface for what is in effect an online storage bucket.   So I guess the answer is “it depends”. 

I was wrong.

24 hours into my first Chrome session it still only uses up 50meg of memory – and stayed at ~20 yesterday for most of the day.  It refused to choke in an IE-like manner over sites like Ebay, YouTube & Gmail.  Flash videos loaded up closer to my PS3 browsers speed (i.e. like something approaching acceptable performance), and as I say, no memory leak.  Internet Explorer would have hogged anything up to 600meg of space by now, thus requiring me to save all my current tabs on notepad/to memory kill it then start the shoddy thing up again.

Granted Gmail and other javascript-heavy sites did not seem to run any faster.  But they didn’t stall either.  It doesn’t come with Java support built in, but I don’t think I’ll risk adding it on – Chrome is my new speedy browsing experience from now on.  Firefox still gets the heavy-lifting vote due to the overwhelming array of plugins I’ve slapped onto it.I’ve been critical in the past towards Google and their useless feature-bloat, particularly in Gmail’s case.  But with Chrome they seem to have done good.

 Update:  Too much hope I guess.  No sooner had I posted this than I found a bug.  Over and above the annoying textbox glitch that seems to be posing quite a problem.  It’s treatment of textbox input is faulty when it sends to the server – It removes newline formatting in WordPress 2.2.1.  Thus my blog post came out as one long blurb.  Bug report submitted.