Unype blog

Javascript is getting very fast through some optimizations. I mean, as fast as C, which is the fastest you can get for a non-assembly language. Firefox 3.1 will ship with this optimized javascript engine: TraceMonkey.

(Graph courtesy of Brendan Eich.)

Paul Graham’s prophesy that web-based applications though slow now, will win at the end when javascript runs at the speed of C, may be coming true

Meanwhile hordes of developers are busy porting their web applications (written in Javascript) into Objective-C (iPhone) and Java (Android).

Why? Because you cannot access the native APIs on these platforms from javascript.

But then all we need is a way to call the native methods on the iPhone and Android from Javascript. And this is already happening at the grassroots level for iPhone as PhoneGap. And the new Mobile Safari DOM model has events for touch/gesture manipulation from javascript. But of course Mobile Safari javascript engine is very slow today, so handling gestures using javascript only works on the emulator. So we are working on our iPhone and Android applications as we speak

But I expect this ‘porting web-applications to native smart phones’ to be a transitory phase. Eventually, to use Facebook, you will just go to facebook.com on your smart phone and will not have to install some application from some ’store’.

Hi5 today announced that they will integrate an embedded virtual world platform that they had acquired earlier this year into their service.

Hi5 is one of the most popular social networks and its userbase is mostly outside the US. They clearly see the value in having their users interact with each other more in real-time, rather than the current bulleting-board-like model where a user posts something and you see when you ‘refresh’ your page.

With a dynamic real-time interaction model like Unype, users do not have to refresh their homepage or relogin to the service to see what is new. Each new action, post and event is propagated to all live clients in real-time. Unype’s backend has been fine-tuned to support just that.

We are looking forward to seeing more social networks develop and employ more real-time interaction models.

We are extremely excited to have my friend Eric Redlinger join Unype, bringing along his deep research and development expertise on mobile devices, especially the iPhone.

We are also excited to present Eric’s immediate contribution: initial screenshots of an early version of Unype on the iPhone:

Eric’s previous projects include work on the popular multiuser media-synthesis platform KeyWorx project and several innovative iPhone applications: DropCopy, a wireless file transfer utility that has already been downloaded over 300,000 times in just 4 months and the shape-shifting MrMr, an amazing iPhone app that can morph its GUI on the fly according to instructions sent from a server.

We are excited to innovate on the iPhone to bring new kinds of user-focused experiences to the device.

Starting a tech company used to be hard. You needed lots of developers, lots of resources, lots of money and more money.

Then web APIs made it easier to start companies. ProgrammableWeb has at last count 861 APIs available for use, and 3257 mashups created using these. Now instead of creating your own map application for your rock concert website, you can use one of 88 mapping APIs available and slap your concert listings on top of that.

In the meantime, there was still one big hurdle for startups: deploying and maintaining a backend that could scale if the startup actually indeed succeeded. Not all startups hit millions of users but still, every startup had to be ready for the onslaught or they could kiss their exits goodbye. This required buying/leasing servers, architecting, rearchitecting and then maintaining clusters of servers, replacing burnt hard drives, backing up, adding more hardware as more customers showed up and keeping an army of sysadmins fed and appropriately caffeinated during the process.

Google and Amazon’s big advantage was their infrastructure. Especially Google’s big competitive advantage was their MapReduce/BigTable-based seemingly-infinitely-scalable infrastructure that could support huge amounts of traffic and data, running on low-cost hardware.

Amazon opened the flood gates to offering this big competitive advantage to any company in the world by becoming the book store that sold cocaine out the back door. As Larry Dignan said “Books will be just a front to sell storage and cloud computing”.

Not a day goes by now without an announcement from another industry giant (Intel, HP, Yahoo, IBM, Verizon, AT&T) offering scalable compute clouds. Dell even tried to trademark the term ‘cloud computing’.

Now that cloud computing is going through what web APIs has gone through, it will be interesting to see if the competitive advantage for startups will be purely innovation now that all startups will be able to scale their service without a huge amount of investment and resources.

The man who gave us Alice, Randy Pausch, the professor at Carnegie Mellon University who also inspired countless students in the classroom and others worldwide through his highly acclaimed last lecture, has died of complications from pancreatic cancer. He was 47.

A must-watch video for anyone, but especially for entrepreneurs (and yes, he was a professor):

* Brick walls are there for a reason: they let us prove how badly we want things and they let us prove our dedication.
* Help others.
* A brilliant quote from his mentor he received as a young man: “It’s such a shame that people perceive you as arrogant, it’s going to limit what you are going to be able to accomplish.”
* You can’t get there alone: always tell the truth, be earnest, apologize when you make mistakes and focus on others.
* Get a feedback loop and listen. When you stop getting feedback, it’s really bad. It means people have given up on you.
* Don’t complain, just work hard.

Tonight’s Cloud Computing panel that I put together was held at the offices of Meetup Inc. (thanks to Jonah Keegan) and had great speakers and a great audience.

Our speakers were (from the left):

Mike Nolet - AppNexus

Jamal Mazhar - Kaavo

Hans Zaunere - New York PHP

Hank Williams - KloudShare (stealth - see Hank’s excellent blog)

Geir Magnusson - 10gen

Ronald Bradford - 42SQL

The panel started with an attempt at defining what cloud computing is. The term has become too popular and hence everyone started using it too loosely. So it now means different things to different people, but mainly, it’s the abstraction of services from hardware and scaling of data services properly so that services can scale to run on any number instances easily for different levels of demand.

But services do not move magically into the cloud and scale. Almost all applications have to be re-architected and rewritten. Engineers in the audience also noted that it requires a rather fundamental change in their thinking.

Different performance requirements almost always require a re-architecting of a service, the panel noted. For example, Ebay had been rewritten from scratch 4 times so far with completely new architectures everytime their performance requirements have changed. Here is a great article about ‘lessons on scalability from Ebay‘

In CC, ‘consistency’ almost always takes a backseat to ‘availability’. That is, it is much more important for a service to be available than for it to be consistent. For example, on Amazon’s SimpleDB, it takes around a minute for a DB write to settle across the system.

Data security is another big issue, and, for example, AppNexus is solving this for some of their customers by giving them dedicated hardware, so there wouldn’t be any ‘jumping out of the VM’ issues.

But eventually not every company with a website wants to have specialized system administrators on their staff 24×7. So, specialization of skills and economies of scale will lead to mass adoption of CC. Jamal noted that CC is now a $200M industry posed to grow significantly over the coming years.

Hank identified 5 different levels of creating a web service:

1) Rolling your own: buying a machine, colocating it, maintaining it, etc.
2) Traditional dedicated hosting.
3) Smart hosting: Amazong EC2, AppNexus, etc.
4) Tools layer: where tools help with the abstraction, for example SQS, S3, SimpleDB
5) Platform: Coghead, SalesForce.com, Google App Engine

Here, 10gen probably is between 4 and 5.

The panel predicted that levels 1 and 2 will eventually disappear or morph into levels 3 and up.

It was agreed that ’standards’ would help spur the demand side but the supply side cannot focus on standards yet since it is still ‘very very’ early in the industry and the innovation cycle, so it’s hard to predict how the industry will evolve and what will become the ‘de facto’ standard that might evolve into industry standard. Cloud computing is a rapidly evolving field that will bring us a lot of innovations and we are excited to watch it grow. Unype will definitely be using one of these solutions to scale.

Along with the new and greatly simplified GUI thanks to feedback from all of you, now Unype also has more ways to connect with others. Here is the latest look:

Here are some of the improvements:

* Simplified GUI. We got rid of the screen clutter to make it easy to get to where you want to get. “Items” and “Tours” have been simplified into the “Shared” tab.

* New ‘quick-access’ latest visitors, voters, contacts, who added me, most popular view. This is a dynamic view of who is looking at you, who is voting for you, who added you. You can also see the list of most popular users and see if you have made it in there!

* Better online users view. The new online users view located to the left of the map gives you a better view of the users that are online in the system. Note that, this view is now ‘collapsible’. You can hide it if you want to see more of the map view.

* Number of profile views! Now you can see how many times a user’s profile has been viewed.

* Popularity! Now, you can vote for users on their profile and also see how popular they are.

We just passed the 160,000 users mark and as always, we are looking forward to more of your feedback, please email us directly at murat at unype.com!

I found this photo at one of the ‘funny things’ websites. And I don’t think it’s funny at all.

Actually I think these people are having a better experience of the internet than you and me right now.

Look at how they are pointing each other to some of the articles on the ‘internet’. Which of us can do it now? They can point an article to each other and then talk about it in real-time. Again, we cannot do it. The most we can do, which is leaving comments on a blog or a BBS would be like leaving post-it notes on this ‘board’ and then leaving without talking to anyone here. Not very real-time or satisfying. These people can have real-time conversations about the content. They are definitely having a better experience than any of us.

On the other hand, for the future Lively, Vivaty, Metaplaces etc are sharing the same vision with Unype: Being able to connect with people in the same ‘virtual’ location in real-time. Unype has a twist, where ‘that virtual location’ is actually a ‘real’ world location. Like Times Square. Or Central Park. Or your backyard.

So Unype can build location-based services on top of its services, as well as ‘permission-based’ location-based advertising. We are building our ’smart’ location-based advertising platform in a way that ‘any’ location-based application developer will be able to plug into it through APIs.

We welcome any location-based application developers out there ( mobile or desktop ) to contact us (murat theatsign unype.com ) and learn more about the platform. It frees you from having to develop a smart location-aware ad-serving platform and at the same time, it would optimize your monetization.

The Symposium on LBS at Columbia University on July 11th was a great one-day event.

Director of the Columbia Institute for Tele-Information, Professor Eli Noam opened the day with 3 questions:

* If LBS is so hot and up and coming, why don’t we have it on Cable and from ISP’s ?
* Is the DB we have today in all these separate companies ( Localeze, etc) complete?
* How will the privacy issues be resolved?
And he challenged the whole group to “Show us” and not “Promise” anymore.

Chair of the Computer Science Department Henning Schulzrinne (one of the inventors of SIP - basis for VOIP)
followed up with a great presentation about IETF’s work on LBS privacy issues: GEOPRIV (Geographic Location/Privacy) and PIDF-LO, a data format that carries with it flags about privacy levels and how long the data can be retained. Privacy levels denote who can access the data (everyone, friends, noone).

Ted Morgan from SkyHook Wireless talked about GPS vs other solutions they provide, namely Wifi positioning and Cell tower triangulation. GPS doesn’t work indoors. GPS requires some time for a fix (getting the latitude/longitude of a location). And GPS drains batteries. He showed technical support pages from Nokia, and some other phone companies where customers are urged to go outside and not move around for upto 5 minutes to get a fix. He also gave a tip for iPhone users: when you hold your iPhone horizontally, the GPS reception degrades a lot. So, to get a fix, keep it vertical at all times. His point mainly was, SkyHook’s Wifi and Cell tower based techniques are a MUST if you want to get a quick fix without draining your batteries.

Evan Neufelf from M:Metrics gave lots of useful numbers in his presentation:
* USA smart phone penetration: 7%.
* USA mobile subscribers: 226M ( France: 46M, Germany: 49M, Italy: 47M, Spain: 34M, UK: 47M, China: 37M - while all US and European markets are saturated, China is such a huge market waiting to explode. Phone companies will need to find new ways to expand their businesses in Europe and USA, while in China, they can still expand their business by getting new customers.

It was also mentioned that when customers were asked what the top factors would be for them to accept location-based advertising, the top response was “It has to be permission-based”, while “It has to be relevant” scored low.

Blair Swedeen from PlaceCast mentioned that customizing the creative for location-based ads ( depending on the location, the customer, the time of day etc) is a big factor in increasing CTR’s. Internet currently is at 0.1-0.2 CTR. They have created campaigns where each customer was given a direct link to the nearest dealership for a car company and a customizable creative ( You are 1.4 miles from a test drive ) where the campaign had very high CTR’s. These advertisers will surely go back to location-based advertising next time they have a campaign.

Ben Ezrick from Ogilvy Interactive has shown a video from MIT about Near Field Communications. He also mentioned that some of the time, they are trying to reach the biggest audience in the cheapest way for their global customers. Reaching people efficiently is a problem. For these campaigns, location-based advertising which provides highly-targeted but more expensive advertising would not make sense. Location-based advertising has to find the right advertiser-audience for itself. And at the same time, for now, this audience is mostly local-businesses. But selling to this audience requires a salesforce, which doesn’t scale. You need a salesforce for every city you want to operate in. So, obviously there are lots of open issues with LB-advertising that will improve in time. There are currently 22M local businesses in the USA. On average they spend $3000 for marketing annually. So this is a huge market. And of course, even though online sales are growing rapidly, in-store sales still account for 95% of sales. Using location-based advertising to drive more consumers into stores represent a huge opportunity for the industry.

Also there are differences between desktop and mobile phone internet surfing. There are no cookies and still no fine-grained tracking of ads on mobile phones. These represent the technical problems that can be solved on the server side.

Overall it was a great day of presentations and brain-storming and I am looking forward to more events like this in NYC.