Manthan

I built a couple of applications using Google Appengine. The google appengine was marketed as highly scalable with high-availability. Google Appengine looked like an ideal choice for “salary search” application that I was planning to build for a while. The data-set that I was looking at was more than 2 Million records to begin with and felt that Appengine could scale well. Two weeks after releasing it, I am frustrated with how little Google Appengine delivers on its promises.

1. Abrupt time-out errors: This is the single biggest issue with Google Appengine. Nobody including Google has explained what causes this and how to avoid these abrupt “Google Appengine is Over Quota” errors. We are not talking “Digg” effect, but something like 3K page views and 100 users per hour. Google appengine couldn’t scale to handle that kind of load.
   • Not doing any writes to the datastore which are more expensive.
   • Every request is memcached
   • Not more than 100 entities are being retrieved per request.
2. 1000 Entity retrieval limit: There is a limit on how many entities of a particular type can be retrieved. For example, when I try to list all the jobs offered by Infosys, the results cannot go beyond a count of 1000. There is a offset parameter but that offset works within this 1000 entities limit. So if the offset exceeds 1000, the query bombs.
3. SDK is broken: The SDk and the production versions are out of synch. What works in the SDK does not necessarily works in PROD version.
4. Indexes: Building Indexes takes more than 24 hours if it is successful.
5. Updating Data is cumbersome: Once the data is uploaded, its very difficult to update that data. There are no “bulk” update operations that can be performed on the data. The only way I am aware of is creating a page and set it to “auto-refresh” using Meta-tag. The data operations functions are so rudimentary that no serious iterative development can be done on app-engine. Small changes to the data-model can take fore ever to implement.
6. Documentation and Support: Documentation is limited. For support one has to be active on google appengine group. Though the engineers on the group try their best, the support needs are lot of improvement
7. CRON/Batch jobs not supported: There is no support for CRON jobs or batch processing.

In conclusion, Appengine is Work in Progress and it has a long way to go before it can live up to its expectations. It is definitely not suited for iterative development, where companies want to release incremental functionality regularly. If the data-model is not well thought out , then Appengine framework will lock all your data or not scale at all, which defeats the purpose.

Making it simple for users to walk away from a Google service with which they are unhappy keeps the company honest and on its toes, and Google competitors should embrace this data portability principle, Eric Schmidt said at the Web 2.0 Conference in San Francisco.

Well to begin with can I get access to my data ? I have no clue want kind of “attention” data is being gathered. The first step to generate confidence would be giving access to the attention data that Google is capturing and a way to manage it. The first step to win over people like me is to inform and share all the data that is being captured ? Where are all my searches sitting, how I can I delete them for ever ? Right now it sounds like Money in the bank, which can be transferred from Yahoo to Google or MSN, but the only person who doesn’t have access to it is the end user. If I delete a mail in Gmail, does it get deleted for ever or it can be retrieved by Google. Search history can reveal a lot and how does one manage/delete it ?

As a loyal Google user and admirer, this blog post by Mark Cuban makes me a cringe. Writing about the details of Google-Youtube deal, Mark Cuban quotes some anonymous sources

> The first request was a simple one and that was an agreement to look
> the other way for the next 6 months or so while copyright infringement
> continues to flourish.
> The second request was to pile some lawsuits on competitors to slow
> them down and lock in Youtube’s position. As Google looked at it they
> bought a 6 month exclusive on widespread video copyright infringement.
> Universal obliged and sued two capable Youtube clones Bolt and
> Grouper. This has several effects. First, it puts enormous pressure on
> all the other video sites to clamp down on the laissez-faire content
> posting that is prevalent. If Google is agreeing to remove
> unauthorized content they want the rest of the industry doing the same
> thing. Secondly it shuts off the flow of venture capital investments
> into video firms. Without capital these firms can’t build the data
> centers and pay for the bandwidth required for these upside down
> businesses.

For me Google is still an upstart with a better product. Takes the market because of a superior product and by putting users ahead of anything else. Stifling competition, getting competitive advantage by evil means ? That’s not the Google I admire and its not in Google’s character to do something like this. I would need more than “unidentified sources” to be really convinced.

Mark Lucovsky and Vik Singh have been fantastic and quick in responding to issues/request with the Google Custom/Ajax search. When released, the customs search had a limitation, whereby one couldn’t use the labels/refinements created on coop in AJAX search. It took them a few hours to roll this out

On Tuesday we announced our support for Custom Search Engines. We asked you for your input and many of you suggested that we add support for Custom Search Engine Refinements. This evening we launched support for Custom Search Engine Refinements in a way that complements what many of you are trying to build. We extended .setSiteRestriction() so that in addition to passing the Custom Search Engine Id, you can also pass a Custom Search Engine Refinement label. See the code snippet below:

var cseId = “017576662512468239146:omuauf_lfve”;
searcher = new GwebSearch();
searcher.setSiteRestriction(cseId, “Lectures”);

The problem I have seen with the custom search is, there is no way one can refine results from BlogSearch. Some of the features I would like to see

• The blogSearch control doesn’t support setSiteRestriction. So the search returns results from all over the blogosphere, with a lot of irrelevant info.
• The AJAX search doesn’t yet support code search. Sorely missed this while creating a custom search for Oracle Applications
• Next/Previous links on search custom search pages, without the need to going back to google home. Right now, you can only show 8 results on the page, but would really like to have navigation to all the results.
• Documentation, not very good and is tough to get started.

Well this is something that I could really use. Searching for Oracle Application related information mostly landed me in “spam websites”. So I ended up using site: parameter for narrowing down the search results. With the launch of “Custom Search Engine“, my job is made simple. There are only a few decent websites/forums which have usable information. But these forums/website have archaic search features. For example forums.oracle.com takes ages to load and brings a lot of irrelevant results. Using Google search for searching relevant information, while turning of the noise from splogs/spams would save me a lot of time. It took me five minutes to come up with my own Oracle Applications Search engine . Sweet.

Hitwise had released figures indicating almost 10% of google search traffic is for image searches. There are no ad’s with Google image search, simply because image search doesn’t provide a good way to put contextual advertising on images. Google’s purchase of Neven Vision can change all that. The official google blog talks about providing a better way to organize personal photos.

We’ve been working to make Picasa (Google’s free photo-organizing software) even better when it comes to searching for your own photos—to make finding them be as easy as finding stuff on the web. Luckily we’ve found some people who share this goal, and are excited that the Neven Vision team is now part of Google.

In the long term, I hope google would be working on ways to put contextual ad’s next to photos. Photos are the single biggest source of user generated content, far ahead of text and video.

It could be as simple as detecting whether or not a photo contains a person, or, one day, as complex as recognizing people, places, and objects

If the image recognition software were to mature, enough to identify objects and places in photos, it gives google an opportunity develop something of an “Adsense for Images”. With the richness of Ajax, these images could be overlayed on images. In the short term, the technology would be put to test with Picassa , helping users to manage their photos better. Gigaom has more on this.

Riya’s Munjal weigh’s in .

What does this mean for us at Riya? It means we now have tangential competition but not yet direct since with Riya 2.0 we are focusing not on personal search but on webwide similarity search

I wouldn’t be so sure, since Google is basically a search company and not a personal photo organizer. Webwide search would be a part of long term goal , which would bring it head on with Riya.

More about the company Google acquired from Google’s cache. The company website has been replaced with a message, its has been acquired by Google.