Monday, July 11, 2011

HTML5 Development

HTML5 is catching attention of many people and companies lately.
Heck, even Microsoft is shifting it's rich client technologies to HTML5...
It's going to take some time until HTML5 will be the mainstream of web development but until then, why not prepare yourself? That's at least my state of mind :)

So here some questions that popup:
Are there any IDEs which already support these new features?
Best practices? TODO/not-TODO
What websites already adopt HTML5 technologies (other than just the

Friday, November 20, 2009

Why youtube is a bad music player

I know, I know youtube is not designed to be a music player...
But wouldn't it be cool if it was? :)
Here some problems that I think prevent youtube from being a music player:

Music search is not powerful enough:
  • Few videos for every song, but finding the best quality version is hard
  • No option to filter out covers, live recordings, deleted videos or videos with no audio
  • No true search or browsing by artist, album, genre, year, etc.
  • No additional lyrics, album art, bio, chords, etc

Poor playback (with playlists):
  • New search stops playback
  • Playing next video can take few seconds
  • Bad videos can jam the playlist
  • Most playlists contain duplicate videos, unrelated , removed or low quality videos.
  • Audio volume varies from video to video
  • Long beginning or ending of audio may be silent
That's it I think... Did I miss anything? Now it's your turn...

Tuesday, July 7, 2009

Finally a Google OS - THE REAL THING!

For those of you who need to catch up:

What an interesting news
Google has taken one of the biggest challenges in the software industry:
- Make a mainstream OS that compete with windows
- Build this OS on Linux (Linux kernel actually)

Now everyone knows Linux is a very solid system, that runs many (most?) servers in the world, but surely not many PCs. Why? Not only because Microsoft pushed windows as a pre-installed system to nearly every laptop! don't buy that!

I love Linux, but I admit it's not for everyone - it has a learning curve that doesn't fit average users. I don't see how my mom, for instance could ever configure a webcam, or the proprietary NVIDIA/ATI device driver. All of these normally require command line magic. Also, try to explain why this damned word document won't display properly in Open Office. Oh, and forget about these shiny 3D games you were playing - they won't work here.

Don't get me wrong - I don't blame the Linux community for these problems. If you insist on blaming somebody, I would blame the hardware vendors (for giving windows such a priority), software vendors (for developing software to windows only) Microsoft (for their Microsoftness), and just then the Linux guys, for not paying enough attention for usability.

Ubuntu and others are trying to improve the Desktop experience, but I've never seen any distro that really solve all of the problems yet.

Well, I don't see how things can be solved that quickly, but Google are indeed a huge force coming in to help. They want to build a zero maintenance system - "It should just work". And to do that, they will first need to convince hardware manufactured to fully support Linux (Software is less important because they target browser-only system). But actually, it might be easier than that, because the vendors will see the hype around Chrome OS and wouldn't want to stay out!

Lets sum up the challenges, and predict how google will face em':
#1 supporting all kinds of hardware - solved without (suprisingly) doing nothing
#2 supporting all kind of software - ignored, just forget about your win apps and run the corresponding app on the web. Problematic, but works for more and more users.
#3 building a "it should just work system" - They will probably do what they did with Chrome - silently update periodically. Stinky if you'll ask a geek, but works.

So Google has undertaken a 500-pound-gorilla-challenge once again, but the odds they will succeed are not bad. For us, the users, it will mean more competition, and better experience. For us, the Linux geeks, it means that Linux is becoming mainstream - this time for real.

Saturday, April 11, 2009

My futuristic web tablet

A few months ago I came across a new project TechCrunch themselves are undertaking. The goal is to build "dead simple web tablet" - that is a light, cheap laptop, with touchscreen for browsing only. Which means the software will only include a light OS and a browser.

That's really a cool idea in my opinion. Lots of people seem to be interested in it, and also got involved. Recent coverage of the project revealed some new photos! I surely want one of these! But I even want more one that can somehow shrink into my pocket. And while being so small, it can be my multi-functional cellphone/camera/GPS a-la-iphone gadget :)

The following is an idea I had about a year ago, and today I've decided to finally share it (It's not that I worked about it the whole time). The basic idea is to cut the screen into 4 parts that are convertible, just like phone are sometimes 2-part convertibles. So here is a little design I just did:

Think about an iphone or even an ipod touch - yes that's better, it's thinner.
Now, we'll make it even smaller, by cutting 3 corners, leaving only the left-bottom corner:

Imagine you're holding 2 of these ipods, one on top of the other. They are also connected on their right sides. Then you can flip (opn) the rear ipod. and get a "double" ipod:

Now what if you actually hold 4 "chopped" ipods, all connected in a way you can make another flip - (double the double)? you'll get this 7" screen which is composited by four 3.5" parts:

In horizontal mode it looks better of course, and makes more sense for web browsing:

Now, it's not 12" laptop like TechCrunch are aiming at, but it's fairly adequate for basic web browsing and email reading. If insisted on 12", we'll need 4 part screen in which every part is 6". But then again, it probably won't fit a pocket. We can go one (giant) step further and suggest 16 parts of 3" to compose the big 12" screen.

Indeed, it's only an idea, and a rough design. Implementing this product involves complex engineering for sure, but if someone into it - let me know! Maybe a possible TechCrunch Pad2? In any case, comments and suggestions are welcome. Thanks!

Monday, January 12, 2009

Blogging VS Twittering

Blogging is pase, Micro-Blogging is IN, and Twitter is HOT.
For those of you who somehow missed the action, Twitter lets you blog short posts, and follow other people that you find interesting (such as Robert Scoble or Lance Amstrong)
Twitter appears in Techcrunch everyday, and I wanted to know what's the big deal about it.

For the couple of months I have used Twitter quite a lot.
At the beginning I was not sure what was the whole noise about...
Then after one month I started to interact more with other people and even gain some followers.
That's the point when I think many people get addicted to Twitter. You are followed, and you are instantly important figure - yeah right... But it does feel good to be followed - I must admit.

Then you get to suddenly following lots of people and you can read lots of posts from them:

Omer Perchik
OmerPerchik sooooo happy! :)
Alex Payne
al3x @pjdoland Brilliant! Videos are good :)
Orli Yakuel
Orli @kevinrose o★m★g it looks like a puppet! LOVE the name too!

And then I realized - WTF am I doing?
I don't have accurate stats on this but about 60% of the posts are conversations, 30% are enigmatic meaningless status updates, and only 10% are really cool stuff worth reading.

Another thing that amazes me is that people follow tons of other people.
For example Robert Scoble follows more than 20000 people, and he's really not that special.
Lots of people follow more than 1000 people which is still a HUGE mass of posts.
It's just impossible to read all the posts from even 200 people. Unless you have more than one hour a day, to read what John thinks about Katie's new profile picture.
It will take a lifetime to read post from 20000, so why do you follow them in the first place.

I really feel like I need a good aggregator that can automagically filter out all these tweets and keep only the good ones (crowdsourcing maybe?). Without a filter, twitter has exausted me, and I'm taking a break... There are more differences between twitter and blogging and twitter is really a phenomena that I still don't fully understand, but I need a vacation.
I would invest my time more on blogging and Google reader :)

Wednesday, December 10, 2008

Playing with disqus

New commenting system...

Sunday, November 23, 2008

Identity Discovery

With so may online services today, there's a need to control them in one place.
Current social aggregators such as friendfeed do that but you have to work for them first.
What do I mean by work? You have to manually add your username for each and every account you want to use! This is a daunting task that can take some serious amount of time.

But why do you have to this manually? There are many identity search engines today that do deep social search in the various data sources. Notably, which brings you astonishing amount of information given just an email field. So why the aggregators don't do that as well? Ask for the email field and discover the user's accounts automagically? Probably something that many devs are working on...

On the same topic, aims to make the life of the aggregators much easier, by giving auto discovery services and API consolidation. Looks great - I'm going to check it out soon :)