Fusion Tables and Shapes

A lot of geographic data in public domain is distributed in SHP format. However, Fusion Tables application supports geographic data only in KML format. Google has recognised the opportunity and is now providing a “translator/loader” application to facilitate uploading of SHP files into Fusion Tables. Shpescape has been implemented with GeoDjango framework and is aimed at facilitating the process of converting and loading that vast resource of GIS data from SHP format into Fusion Tables – to improve uptake of Fusion Tables by GIS as well as broader application development community.

The concept behind Shpescape is great but it fails in terms of performance. I tried the application with a modest size SHP datataset (40MB) and the result was less than satisfactory. It took extremely long time to upload the data to the server, process it into KML and load into the Fusion Tables (short of an hour!). I know from my own experiments that converting SHP into KML takes only a few seconds with basic PHP script. Allowing for download and upload time (since 2 separate servers are involved), the whole process should be finished in a matter of minutes and not almost an hour. The biggest disappointment was that the algorithm used in Shpescape enforces generalisation of polygons and does not process “point for point” from SHP to KML. It resulted in some polygons being converted incorrectly and/or corrupted in the process (as per image below).


Shpescape may work with small SHP files, with simple geometries but, as it stands, I do not recommend using the application with full resolution datasets. Rather, use alternative SHP to KML converter and download KML files directly to Fusion Tables.

Via:AllThingsSpatial

Google Hotel Finder Experiment

Last week amidst the noise of changes in the Place’s layout, Google noted  that  they would be “Integrating some of the great information that’s been buried on Place pages into your web search experience across all Google platforms“. While Google doesn’t talk much about their thinking or the future, when they do, I have learned that you can take them at their word. They have in fact quickly started this process of projecting data from within Places more broadly, resulting in its higher visibility and an increased liklihood of being seen. They are putting more photos from Places into the branded One Box in the main search results and have started showing coupons from Places in their mobile Shopper app (OMG will coupons finally make their Phoenix like  re-appearance after 4 moribund years?) .

The Google Hotel Finder experiment is yet another example that they are taking buried information from Places and Maps and experimenting with making it more visible. In the process they are making search results more engaging and interactive, demonstrating their move from being strictly a search results provider to using search to generate useful content that will attract and retain users. More users, a longer time on the site and a fresh way of looking at previously buried content will obviously also provide additional ways to sell more ads.

Rather than the standard Google approach of the single search box and their educated guess as to what searchers want, the Hotel Finder interface, in very un-Google like fashion, provides a more faceted approach to finding exactly the information that a user is looking for. The choices allow for a great deal of granularity of pricing, relative pricing and quality.

If the broad, single field geo search does not return the appropriate geography, the user can drill into the map and literally outline the appropriate neighborhoods themselves via interactive, draggable boundary lines. The map view provides a heat mapped representation of the most popular areas.

The interface allows a user to build repeated queries with slightly different parameters and save the results into a “short list” of choices. Thus if you wanted to compare hotels in two or three totally distinct non-adjacent neighborhoods, say the Upper West Side, Tribecca and Park Slope, a user  could create a custom view of hotels from which to choose and then share the view via URL with another.

More details about a given hotel, a Places view if you will, with a very attractive layout can be seen by clicking on the hotel of choice. The user is presented with an array of photos, review summaries and the owner description. It seems to reflect a new, thoughtful design sensibility on the part of Google.

Of course, it is not just a view of content but offers the option of booking the hotel via their still secretive hotel booking tool. All in all it is in impressive experiment with a subtle transactional nature. It is both more polished and definitely better looking than most Google experiments. It is very slick and offers an interface that could be easily adapted to restaurants, bars, florists and hair salons (to name a few) and of course to mobile.

If this experiment is any indication, the future of search is local search and it is an interesting one. With the acquisition of ITA and Google’s obvious and long standing desire to move into the hotel booking market, this experiment shows how Google is thinking about both the data and the market. Many have explained the recent changes as a reactive response to anti-trust complaints. It could equally be explained as a proactive measure that would allow Google to be in a more competitive position going forward as they compete more directly against the likes of TripAdvisor and Yelp.

The Google Hotel Finder experiment is not just search as we have come to know it but search as interactive content that has the ability to achieve serendipity in both interaction and results. And of course in a way that makes the sale.

Google Places: Reputation Management

Most small businesses live in dread of the day when a competitor drops a nasty review on their Places page. Imagine waking up one day and finding 58 of them. That’s what happened to the Place Page for Moishe’s Moving Systems in NYC. For several days in early July they were finding one 1 star review after another showing up on their Places page. Imagine their sense of futility as they hit the “flag as inappropriate” link over and over again.

A quick call to their competitors across town indicated the same was happening to them. Not just the same pattern but the very same reviews, same bad English, same mispellings, often not even getting the company name correct.

A search in Maps on the phrase “It really hurt me and I highly recommend that NOBODY DO BUSSINESS WITH THIS COMPANY>>>>>> and by the way all the locations they advertise with are 100% fake” surfaced the very same reviews on over 100 moving companies country wide from Miami to LA.

It seems that in this scam, hundreds of moving companies across the U.S. not only ALL received the exact same bad reviews but many then soon received unsolicited proposals to “remove malicious, old, slanderous, unfounded, and internet defamation ratings”.

 

The internet has spawned a whole new generation of reputation management firms that help make sure that the front page of Google does not have bad things prominently displayed about your company. With the growing importance of reviews and the impact that they have had on businesses a number of companies jumped into the “positive review” only game to be sure that your Places page showed only glowingly satisifed reviews.

But apparently, the review reputation management business has taken on a new, more sinister twist of late. It appears that unscrupulous “reputation management” firms are now not only offering to place postive reviews on your Places page and help take down negative reviews, they are actually creating the negative reviews in the first place. Now that’s a business model! Have you seen this practice in other industries?

Google has indicated that they are in the process of removing the reviews. That being said it does highlight the structural problems caused by a still immature review spam algo AND the frustrating process for an SMB to request that a review be removed via the “flag as inappropriate” link. This problem is much like the issues that they confront with bugs in the Places Dashboard process.

It is likely that this obnoxious review spam will be taken down, it is also likely that the spam review filter algo will improve over time.

The current automated flagging system however is inadequate to handle the situation until such time as the algo improves. The flag, like many Google complaint processes, is likely just feedback to their machine learning system. It rarely if ever leads to an immediate takedown. Google consistently prioritizes their needs and the assumed needs of reviewers in this process. It certainly provides NO feedback to the affected merchant as to what if anything Google will do about the problem review.

As demonstrated once again by the review snafu last week, when numerous revews were lost, reviews are an very much a flash point for most business users of Places. The lack of quick public response on Google’s part demonstrated either an incredible lack of staff, an incredible lack of sensitivity or perhaps just an on-going tin ear to the needs of their small business clients vis a vis reviews.

Until such time as the algo is significantly improved and problems like extortion spam can be greatly minimized in an automated fashion Google needs to create a process that comes down in favor of the SMB reporting the problem. Perhaps one that hides the egregious reviews pending a human review process that actually includes timely communication. Once the algo has been refined they could then think about a cut back to the human intervention.

But with Google’s growing portfolio of Local Commerce products they will find a very chilly reception indeed on Main Street until they do a better job of handling reviews.