All posts by user

Science project: does spam attract spam?

We get a ton of spam on this blog, all from robots so they appear in waves: dozens of similar sounding comments that are attached randomly to various blog posts.

A little while ago we decided to run a little science experiment: we took one of the funnier spam comments (they are all inadvertently funny, since they are composed by people who have a shaky command of the English language), and decided to reuse it as blog posting.

The science experiment was to see if a spam blog posting would attract the same amount of spam as a real blog posting: in other words, to test whether the spam engines made any attempt to read the content of blogs before providing their inane commentary.

The results are in: spam blog postings attract as much spam as real blog postings.

Goodbye, Internet, it was nice knowing you…

(And, our thanks in advance to all the spam engines that will respond to this blog posting: keep up the unintended humor; it makes us feel superior.)

The “S” stands for Siri, for speech.

Apple just announced a new platform and nobody noticed.

Far too much of the commentary from Wall Street about the launch of the iPhone 4S has been superficial, focusing on the fact that it is a “4S” and not a “5”. Why, oh why, wail the analysts, couldn’t Apple just have called it the iPhone 5 and made everyone happy?

This misses a rather big point about the iPhone 4S: the “S” may ostensibly stand for “speed”, since the new phones have a faster processor, but in our opinion the “S” really stands for Siri, as in “speech”.

OK, so we don’t have our hands on a real iPhone either, which means we are guessing, too, and using the very limited collection of videos and demos that are publicly available for our guesswork/analysis, the most extensive of which is Apple’s own promotional video.

This promo video, however, does provide a very good indicator of the vast potential of the new speech platform that Apple has just launched. Yes, there’s a whole new platform for personal computing out there now, thanks to Siri, and it’s not just limited to iPhones.

Siri is more than speech processing: it is an impressive attempt to bring interactive voice response (IVR) to the masses in a way that does not immediately induce rage against the machine. There are two major innovations with Siri, only one of which has gotten any real attention:

  • There is the AI needed to understand the user’s speech and translate the nouns, verbs and indirect references (like “this weekend”) into API calls, and here Siri promises to be a considerable leap forward from the kind of hugely annoying systems we all accustomed to dealing with when we call our bank or insurance company.
    There is also the AI needed to generate voice output from the system, and here Siri looks like it could do with some work, particularly if it is to sound less synthetic.
  • The far more impressive work is the Siri’s integration with the calendar, email, Google Maps and iTunes programs, as well as external web services like weather.com. This is the new platform that just got launched: the integration of a speech interface to all personal computing devices: not just phones, but also tablets, laptops and TVs.

The iPhone 4S will debut with Siri working seamlessly with mail, calendar, Google Maps, SMS, weather, and basic Web searches. But this is just the start: what Siri does today is similar to what the iPhone 1 did on the day it launched, when it came with a handful of standard applications for email, calendar, SMS, weather and basic Web searches. (Hmm… sound familiar?)

If the iPhone 1 was impressive right out the box, before hundreds of thousands of applications had been written for it, imagine what the Siri platform will be like when there are hundreds of thousands of existing applications and software packages that support a speech interface.

Apple has launched what they hope with be the speech platform for personal computing. Sooner than we developers would like, we will have to start supporting a speech interface to our personal computing applications, and it will have to be a much smarter interface than anything that came before.

If Apple succeeds in making the Siri interface a de facto standard, they will own a fundamental platform technology that would make Bill Gates envious.

Like a project? Copy it with a single click

We are encouraging our users to create public projects: projects that are open to the public to visit, and which can serve as templates for other users who need to perform similar functions. For example, a template on how to use Michael Porter’s Functions of a Competitor Intelligence System, adapted from his classic book on Competitive Strategy, might be useful for a number of people who are either studying the topic as part of an MBA program, or are simply applying these concepts as part of their jobs in marketing, product management or strategy.

Whenever you visit a public project, you now have an easy way to make a personal copy of that project using a new “Copy Project” button that we have introduced. Clicking on this button makes a complete copy of the public project for your personal use: it gets added to your own Account, and you can start working on it for your own purposes right away. This is what it looks like:

The Copy Project button lets you make a personal copy of a public project
The Copy Project button lets you make a personal copy of a public project

If you like the idea of contributing your ideas to the Kerika community, please consider making some of your projects accessible by the public. If you have a methodology, process or set of ideas or content that you want to share with the world, it’s easy to make your project accessible by the public: just click on the Team button and check the “Make project public” box, like this:

Make your project open to the public
Make your project open to the public

Remember, when you make your project open to the public, it doesn’t mean that anyone can just wander by and vandalize it. It just means that people can view your work, and if they like what they are doing, they could ask to join your project (which is great if you want to attract people to an open source or advocacy project), or they could use the “Copy Project” button to make a personal copy of the project that you have so kindly provided!

Our latest version, and then some!

In the immortal words of Jim Anchower: “Hola, amigos. I know it’s been a long time since I rapped at ya.”

Our apologies for not posting blog entries for a while, but we have the usual excuse for that, and this time it’s true: “We’ve been incredibly busy building great software!” It’s going to be hard to summarize all the work that we have done since June, but let’s give it a shot:

  1. We have curved lines now. And not just any old curved lines, but the most flexible and easy to use drawing program that you are likely to encounter anywhere. You can take a line and bend it in as many ways as you like, and – this is the kicker – straighten it out as easily as you bent it in the first place. There’s a quick demo video on YouTube that you should check out.
  2. We have greatly improved the text blocks feature of Kerika. The toolbar looks better on all browsers now (Safari and Chrome used to make it look all scrunched up before), and we have added some cool features like using it to add an image to your Kerika page that’s a link to another website. (So you could, for example, add a logo for a company to your Kerika page and have that be a link to your company’s website.) Check out the nifty tutorial on YouTube on text blocks.
  3. You can set your styling preferences: colors, fonts, lines, etc. Previously, all the drawing you did on your Kerika pages was with just one set of colors, fonts, etc., but now you can set your own styling preferences, with a new button, and also adjust the appearance of individual items.
  4. We have improved the whole Invitations & Requests process. Now, when you invite people to join your projects, the emails that get sent out are much better looking and much more helpful, and the same goes for requests that come to you from people who want to join your projects, or change their roles in your projects. Check out this quick tutorial on how invitations and requests work.
  5. We have made it easier for you to personalize your Account. You can add a picture and your own company logo, which means that when you use Kerika your users see your logo, not ours! Check out this quick tutorial on how to personalize your Account.
  6. We have hugely increased the kinds of third-party content you can embed on your Kerika pages. The list is so long, we really should put that in a separate blog post. We have gone way beyond YouTube videos now; we are talking about all the major video sites (Vimeo, etc.), Hulu, Google Maps, Scribd and Slideshare… The mind boggles.
  7. Full screen view of projects. There’s a little button now, at the top-right corner of the Kerika canvas: click on it and you will go into full-screen mode, where the canvas takes up all the space and all the toolbars disappear. This makes it easy to surf pages that contain lots of content, or work more easily with your Google Docs.
  8. Full support for Internet Explorer 9 (IE9). Not as easy as you might think, given that Microsoft has historically gone their own way, but we have sweated the details and now Kerika works great with IE9. As Microsoft continues to converge around common standards, this should get easier for us over time.
  9. Full support for all desktop platforms. OK, so this isn’t really a new feature, but since we are bragging we might as well emphasize that Kerika works, and is tested, to work identically on Safari, Chrome and Firefox on Windows 7, Mac OSX and Linux.
  10. Literally hundreds of usability improvements. Yeah, okay, we should have gotten it all right in the first place, but our focus over the past few months has been very much on working directly with our early adopters, observing them use the product, and noting all the tiny friction points that we could improve upon. We are not saying that we have all the friction removed, we are just bragging about the hundreds of tweaks we have made in the past 3 months.

Since June, we have had two major releases: one at the end of July that had nearly 150 bug fixes and usability improvements, and one this week, with over 120 bug fixes and usability improvements.

The product is now in great shape from an infrastructure perspective: the core software has been well debugged and is now very robust. Performance is great: you should get sub-second responsiveness when working in an environment with decent broadband wireless, where you see updates to your project pages in less than one second after a team member makes a change. (We test this with users in Seattle and India working simultaneously on the same project.)

Having this robust infrastructure that’s been well debugged and tuned makes it easy for us to add new features. In the coming weeks, look for more social media hooks, a revamped website, an extensive collection of public projects (that you can use as templates for your own work), and more. Much more. After all, if “less is more”, just think how much more “more” could be ;-)

Some thoughts on software patents

A very big day for us at Kerika: after more than six years of perseverance, we finally got our first patent issued today: United States Patent No. 7,958,080. (Yeah, that’s certainly a big number: a lot of people were granted patents before us!)

There is a lot of debate in our profession about software patents: strong opinions have been voiced in favor of granting software patents, and equally vociferous views have been aired in opposition to the entire concept. What’s discussed much less frequently is how long it can take to get a software patent, or expensive and arduous the process can be.

Our original provisional application was made on October 29, 2004; this was followed up with a final application, and then dead silence for a couple of years. When the Patent Office finally examined the patent application, there was a multi-year exchange of correspondence with the Examiner, with the Examiner making various arguments in opposition to granting a patent, and our lawyers making counter-arguments in favor of Kerika being granted a patent.

All this back-and-forth is quite normal, by the way, and works as designed: it’s the Patent Examiner’s job to not grant patents without challenging their claims, and it’s the applicant’s responsibility to offer increasingly stronger arguments in favor of being granted a patent. What is finally granted is usually a subset of the claims that were made in the initial application. It’s very rare for all of the initial claims to be granted.

Software patents are harder to get, and take much longer to be granted, than hardware patents, probably because it is easy to compare one tangible object with another and determine whether one is materially different than the second tangible object.

Software is harder because it is intangible: you have to describe an intangible item, using the jargon that’s peculiar to patent applications, in way that clearly distinguishes it from another intangible item.

The net result is that a software patent can take many more years to be granted, in comparison to a hardware patent. One of the problems with the process taking so long is that, by the time the Patent Examiner gets around to examining your application, the technology may have become so prevalent in other, competing products that it no longer appears to be as innovative as it was when the application was first made.

The entire process is also very expensive, particularly if you are a solo inventor or small startup (as we are): the Patent Office has been hiking its fees over the years, with the stated intention of hiring up more examiners, and thereby speeding up the process, but clearly they are having a hard time keeping up with the flood of applications because it seems like the delays are getting worse, not better.

Regardless of whether you are pro or con software patents, we would argue that both sides of the argument would be served better if the process of examining, granting or rejecting a patent were shortened very considerably. If a patent could be examined and acted upon within 2-3 months, it would entirely change the debate as to whether software patents are good or bad for innovation.

Meanwhile, back at the ranch… we have other patent claims in the works, which we hope will take less time to work its way through the Patent Office now that our first patent has been granted!

Our latest, greatest version: Copy/Paste, Share/Join, and more

We have been hard at work over the past couple of months: taking in feedback from our early adopters and using this to fix bugs, improve usability, and provide some very useful new features.

Here’s a quick summary of some of the stuff we have built in the past month or so:

  1. Added Copy, Cut, Paste as New, and Paste as Links. The Paste as Links feature is particularly powerful.
  2. Added a Share! button and a Join! button that makes it easy for people to create public projects and get people to join in their particular advocacy cause.
  3. Brought back double-clicking (after having taken it out in an previous version…)
  4. Made the text blocks more flexible, with the ability to embed URLs and autosizing.
  5. Improved our Google Docs Finder.
  6. Worked around numerous limitations of Google Docs.

We will describe each of these in more detail in subsequent blog posts, and we obviously have a lot of work on our hands to update our website to reflect all this new functionality! Oh, and about 200 bugs were found and obliterated in the past 2 months…

(Lots of people have provided useful feedback, but special thanks go to Michael Parker from Marketingeek and Alexander Caskey from High Bridge Communications.)

 

Running Windows on my iPhone…

Some years ago, a friend presented me with an old iMac that he didn’t need any more – one of those blob-like machines with a clear cover that let you see its innards. It was a heavy, inconveniently shaped machine that nevertheless started a slow migration to all-Mac household today: Mac laptops, iMac desktop, Mac wireless router, iPod, iPhone (and, soon, iPad).

Macs were always attractive for a simple reason: you didn’t have to worry about tuning the operating system. When I had PCs at home, I found myself having to do a clean re-installation of the operating system every 9 months or so, just to clean out the grit and gum that clogged up the Windows registry and system directories.

The way Windows programs were installed was particularly problematic: each application would leave bits and pieces of code and configuration files lying around all over the place, where it would inevitably trip up some other program. If you liked trying out different software, as I did, this meant that every 9-12 months you had to rebuild your PC from scratch: repartition the hard drive, reinstall the operating system (and several years worth of updates), and then all your applications, and then all your files. It typically took a full day, but it was worth it because your machine was all zippy and fast once again – at least for a few months…

The main selling point of Macs was that all this fiddling with your machine was unnecessary: your computer would “just work”, and you wouldn’t have to worry about rebuilding the machine periodically to get it in fighting trim. And, with the new iCloud strategy, Apple is trying to take that brand promise even further, by demoting your expensive gadgets to “just another device”.

Is this all going to work like magic, as Jobs promises? I doubt it, based upon my recent experience with my iPhone… For the past few months I had noticed that many of my basic iPhone applications would just quietly crash when I first tried to launch them: launching the music player, for example, would frequently take two attempts. The maps application would take up to a minute to launch, and then it would operate so slowly that there would be a 2 second delay in echo-back of character input.

I assumed this was due to the fact that the iOS operating system was getting fatter with every new update, while my hardware remained svelte. After all, the upgrade to iOS 4.0 had been particularly painful for my iPhone 3GS: it taught me a painful lesson about not automatically taking the latest software updates, even from Apple!

(And it wasn’t just me: lots of iPhone 3GS users complained bitterly after upgrading to iOS 4 – it was like trying to run Vista on hardware designed for Windows 95)

And that’s the long-winded segue into why it feels like I am actually running Windows on my iPhone: a couple of weeks ago my battery started to run out very quickly (it wouldn’t last even a few hours of slight use), I took my iPhone to the local Apple genius bar where they quickly determined that my iPhone’s OS was in terrible shape. Their recommendation was to do a complete rebuild of the iPhone: from scratch, which meant setting it up as a brand new phone and adding back in all my email accounts, music, etc. by hand.

This rebuild worked: now my iPhone is as fast and responsive as it used to be, but the experience of completely rebuilding an iPhone (which, incidentally, is not the same as doing a “restore from backup”), was all too reminiscent of having to rebuild my Windows machines periodically to clean out the gunk in the registry and system directories.

It is to the credit of the Apple “geniuses” that they quickly determined that my phone’s OS was in trouble, but I also suspect it is because this is a “known but not publicized” problem with iOS 4. Over time, iOS 4 gets in trouble just like Windows XP and Vista (and perhaps Windows 7 as well, although I use that too infrequently to say for sure).

Think different, work same.

Out of the billowing smoke and dust of tweets and trivia emerges our first patent

“The literati sent out their minions to do their bidding. Washington cannot tolerate threats from outsiders who might disrupt their comfortable world. The firefight started when the cowardly sensed weakness. They fired timidly at first, then the sheep not wanting to be dropped from the establishment’s cocktail party invite list unloaded their entire clip, firing without taking aim their distortions and falsehoods. Now they are left exposed by their bylines and handles. But surely they had killed him off. This is the way it always worked. A lesser person could not have survived the first few minutes of the onslaught. But out of the billowing smoke and dust of tweets and trivia emerged U.S. Patent No. 7,958,080 to be issued on June 7, 2011. Once again ready to lead those who won’t be intimated by the political elite and are ready to take on the challenges America faces.”

Out of the billowing smoke and dust of tweets emerges our first patent
Out of the billowing smoke and dust of tweets emerges our first patent

Our first patent! We are hoping the coming apocalypse doesn’t cause any more delays at the Patent Office; we have been waiting a long time for this patent to get issued, and there are more in the pipeline!

Nice discussion here.

What’s your best guess about the following? Was it written by a computer, written by a human in a different language and then poorly translated into English, or written by a steering committee with representatives from all the relevant business units and impacted stakeholders?

Nice discussion here. I want to join with you by giving my comments in your web page. First of all, your page and design of your website is so elegant, it is attractive. Also, in explaining your opinion and other news, frankly you are the best one. I’ve never met other person with special skills in my working environment. By using this comment page, you are great to give me opportunities for announcing some events. Secondly, I’ve checked your website in search engine result page; some of your website pages have ranked very well. Additionally, your topic is relevant with my desired one. Lastly, if you don’t mind, could you join with us as a team, in order to develop website, not only general web page, but also many specific websites we must create for our missions. From time to time, we, I and my team have great position for you to develop that.

Yup, we really are the best one. We hope this topic is also relevant with your desired one.

Safe house, honey pot, or just dazed and confused at the Wall Street Journal?

The Wall Street Journal is creating its own version of Wikileaks, call SafeHouse, where you are invited to

Help The Wall Street Journal uncover fraud, abuse and other wrongdoing.

What would they like from you?

Documents and databases: They’re key to modern journalism. But they’re almost always hidden behind locked doors, especially when they detail wrongdoing such as fraud, abuse, pollution, insider trading, and other harms. That’s why we need your help.

If you have newsworthy contracts, correspondence, emails, financial records or databases from companies, government agencies or non-profits, you can send them to us using the SafeHouse service.

“SafeHouse”, however, sounds more like a “honey pot” when you read the WSJ’s Terms and Conditions, which offers three ways of communicating with SafeHouse:

1. Standard SafeHouse: […] Dow Jones does not make any representations regarding confidentiality.

2. Anonymous SafeHouse: […] Despite efforts to minimize the information collected, we cannot ensure complete anonymity.

3. Request Confidentiality: […] If we enter into a confidential relationship, Dow Jones will take all available measures to protect your identity while remaining in compliance with all applicable laws.

OK, so as they say, “You use this service at your own risk.” But, here’s where things start to get puzzling:

You agree not to use SafeHouse for any unlawful purpose.

Huh?

If you upload or submit any Content, you represent to Dow Jones that you have all the necessary legal rights to upload or submit such Content and it will not violate any law or the rights of any person.

How can anyone provide confidential documents that belong to another organization without violating the rights of that organization?

Even if you put aside for a moment the absurd notion of a “SafeHouse” where you can post materials that belongs to others, without violating the rights of these others, or any laws for that matter, there is a more puzzling question of why the Wall Street Journal has decided to create their own version of Wikileaks in the first place.

An editorial from June 29, 2010 was entitled “Wikileaks ‘Bastards'” which effectively summarizes their views on anonymous leaking. The WSJ has been consistent in their stance: there are other editorials from April 25, Jan 21, Dec 31, etc., and even today’s editorial page notes that

…we have laws on the books that prohibit the unauthorized disclosure of government secrets. Those laws would fall hard on any official who violated his oath to protect classified information.

So why would they try to create their own version of a communication forum that they clearly despise? An editorial from Dec 2, 2010 provides a clue:

We can’t put the Internet genie back in the bottle.

Well, one place the Internet genie hasn’t yet visited is the Wall Street Journal’s main website: a search for “SafeHouse” there yields no results. Nothing to see here, folks, move along now…