I've been quiet, too damn quiet, for far too long. You see, as the Dude would put it, "lots of shit came to light" and when it hasn't been terrible, it's been so bad that blogging has been at the very bottom of my priority stack. But not tonight, oh no. Tonight I have an ax to grind, and I'm going after cloud computing.

The cloud computing concept nowadays breaks down in two major camps:

• that of the Storage, and
• that of the CPU (or virtual machines if you prefer)

Cloud storage just works and we love it. Someone gives us a REST/SOAP interface and we tell it "here's this blob, store it under this key", we get "kthxbai" and we go our merry way. This is the perfect image of cloud computing. There's no managing of storage, you don't have to request "extra" disk space or "manage" partitions. It Just Works (tm).

Virtual machines on the other hand, sorry! This is not cloud computing! At least to me it's not. Having to create system images? Manage the full software stack? Networking? If this is a form of cloud computing then bow to IBM, HP and whatnot because have been doing it for ages as they manage the mainframe hardware and your sysops (which at this point should have Cloud Computing Expert on their business cards) manage the OS and everything that runs in there. You don't buy hardware, you lease it! So all of the sudden it's cloud computing because it's all in datacenters you have no access to and you pay a lease?! Give me a break! Yeah, this new wave of cloud computing experts/consultants are just glorified sysops, as much as flight attendants are just glorified waitresses. At least the latter actually do business in the clouds, but I digress. Anyway, the only innovation I see on this front is on demand resource availability. (Almost) all the rest: crap!

It's not all bad though, I can see light at the end of the tunnel and it comes in the form of Google App Engine... and something else (more on this at the end). It's not the only solution mind you but it's by far the most popular and most well executed.

You see, in GAE you don't have to manage system images, you don't have to manage networks and all that jazz. You deploy to it, give it instructions on how the world can interface with it and you're done. You don't know or care how many CPUs you're code is running on, where they're at and how to reach them. It all happens under the hood, a hood made by clouds. And let me tell you: it is beautiful! Go try it yourself if you haven't done so already and witness how awesome it is. Don't let the snake bite you, you RoR fanboy, the word is that now that GAE drinks Java, JRuby can also deploy to GAE so go create your awesome 20 minute blog with RoR and give it a try.

But GAE isn't the end solution, oh no, there's something even better, way better, down the pipeline and it's called <your browser name goes here>. You see, a group of smart people hooked up and 9 minutes later this idea that browser clients could be used to process chunks of work came to birth in the form of collaborative map/reduce. Fortunately I had the privilege of being introduced to this idea a few weeks earlier so it wasn't really new, but I was left breathless by it's simplicity and a whole new world of possibilities came to me and I just hope I have the will power to bring at least one to life.

And this, this is cloud computing. Millions and millions of workers requesting serialized code and data they can munch on and you won't have to do a thing to make them work for you. Beautiful! Awesome! And it's light years ahead of this stupid concept that of having to manage virtual machines and virtual networks is the best thing since sliced bread.

Wonderful article by Paul Golding! If you're thinking about
startupping in the mobile market, read it and leave the whishful
thinking at the door.

http://feeds.feedburner.com/~r/WirelessWonders/~3/465199034/RSSRetrieve.aspx

PS: as you may have noticed I'm doing low text, ugly links, kind of
posts. I'm actually using the mail2blogger feature, but I'm not happy.
Need to cook up something that understands the most basic of markups.

Check the link below. For years my preachings on how good synaptic is/
was have been hushed by moronic PC users and now they're going apeshit
with AppStores. I predict a million of them will popup and quickly
become irrelevant, just as download.com did before them, due to all
those thousands of apps that do the same thing, all look alike but the
names are different.

If you think the Apple AppStore is in any way successful (still
waiting for the real numbers) then it's because it's *exclusive* and
*targetted*.

Please, someone, anyone, put a stop to this before it gets out of hand
and those poor users have to install tens of AppStores on their poor,
little and gray and dull PCs to download those copies of shareware
software they don't really need. Unless one of those AppStores
supports illegal torrents, then we're talkinng business.

http://www.techcrunch.com/2008/11/25/ycombinator-startup-creates-a-better-download-app-store-for-windows-baseshield/

-
Celso Pinto // 7syntax // http://handivi.com

This year's codebits edition is over and it finished with the promise of it taking place yet again next year. I think I said it last year but I'll repeat it again: it was awesome! Congratulations to everyone at SAPO for pulling it off. One thing I noticed was that the audience was (or seemed to be) younger than last year's edition, which can only be a good thing.

I'd also like to thank SAPO for accepting the applications for everyone at 7syntax and inviting us to give a couple of talks. VD was there talking about AWS and I gave an overview of the tech we're using to build handivi. My slides are available on slideshare so you can view them if you want:

My project, a topic detection tool, didn't win anything but I'm cool with that, in the process I found out someone was working on a similar problem so, in spite of wanting to stay at home due to coming down with a flu, I ended up showing up for the last day, presenting my creation and being introduced to the Programming Collective Intelligence book which is awesome too. Thank you Paula!

But that wasn't my only hack :-) A few days before the event I had challenged everyone at 7syntax to show up with a mustache. Alas, only Nuno and me did that. Thanks to Alexandre my looks for the duration of the event have been made timeless on flickr:

Pretty out of whack stuff, but cool nonetheless. All I wanted was to have a 60's UNIX hacker look, instead I was called a mexican, a drug lord, a 70's porn star, a heavy metal fan, a chopper rider and James Heltfield-clone. Daniel was really keen on the mexican thing and he was supposed to get me a sombrero for me to wear during my talk but didn't. I would use it, definitely!

Another cool thing was Rock Band, man the game's amazing! I teamed up with Mário, Paulo and a guitar player who shall remain nameless because he doesn't want the internet to know who he is ;-) to play a couple of oldies: Detroit Rock City by KISS and Run to the hills by Iron Maiden. I had to bail out on Friday before the battle of the bands contest due to the flu worsening and becoming exhausted from having only slept about 1:30, 2 full hours the night before but I heard they did well even though some other band won the contest.

Finally, I had the pleasure of spending some time with the massive from up north and meeting Jan and Peter face-to-face.

Combining Codebits and SHiFT it's easy to see how SAPO and it's people have already done more, way more, than anyone else in fostering a creative environment in Portugal. Hats off to them once again!

http://gigaom.com/2008/11/04/google-wins-big-at-fcc-today/

These are great, no, AWESOME news: the FCC allows independent use of
the freed TV wireless spectrum and the deal between Clearwire and
Sprint, paving way for an US national-scale WiMAX network. I know many
of you don't see any future in this so please let me paraphrase a
famous quote on Wayne Gretzky[1]: Wayne became the best because he
skated to where the puck was going to be, not to where the puck has
been.

I'm sorry for the terrible formatting but I'm catching up on the news
in bed and posting this through the iPod email client.

[1] if you don't know who Wayne Gretzky is then you clearly didn't
enjoy enough faceoff time in NHL94 :)

Hugo made some stickers this weekend and brought them in today. As you can see by the photo above, I have plenty of visible area where I can put them so it dawned on me that I should also carry around logos of other local startups, if only to spread the word. So, if you're a Portuguese startup: please send me an SVG of your logo and we'll try to get them printed and I'll proudly put them on my laptop screen. Email to cpinto at 7syntax dot com ok?

Also, I'd like to thank Hugo (aka htr) for doing this. Instead of having fun coding in Python and fiddling with CouchDB he spent an hour printing stickers. Rock on Hugo! ;-)

So it seems that Google's now able to index PDFs by extracting text from images. If you ask me, this is pretty exciting stuff although I have a hunch they're just eating their dog food and, if so, the technology isn't new, it's just Google being Google (ie. smart). At a point where everyone appears to be thinking about semantic search and other (mostly useless at this point) non-sense Google takes it one step beyond and continues to improve their service. This, in itself, is begging for the question: has the Google competition called it quits on the search engine marathon?

It's a slow weekend so I'll take the time to pick the meme where José left it:

1. wake up early: for some reason I keep going back to getting up at about 9/9:30AM when what I'd like to do is get up at 7, 7:30AM maximum. This alone should improve my QoL dramatically as I'd have lots of time for the next item. I'm working on it though :-)

2. exercise more: I have all this exercising gear at home but usually I only stick to the exercising plan for two weeks, then leave it. I should get me a proper bicycle and enjoy early mornings by going out for a ride.

3. travel often and longer: my few latest trips to other countries end up being no longer than 4 days which totally ruins the experience. On the arrival and departure days there's no time to enjoy the time being spent there so this leaves me a couple of days filled with rushing everywhere with no time to let the culture sink in. I guess it's one of the bad things of living at the far tip of Europe: travelling anywhere needs too damn preparation.

A good debunking of iPhone sales by Tomi: http://communities-dominate.blogs.com/brands/2008/10/so-iphone-reach.html

If the link is broken I apologize, as I'm testing the mail2blogger
thingy straight out of byline.

SHiFT, a conference about "social and human ideas for technology" ended last Friday after three days filled with creative thinking. All the team was there and not only that we managed to take one extra lucky person with us, judging by her comments and grin on her face, I'd say she had a wonderful time, which is awesome.

This was the second edition of the conference and not only the big things were done right, all the little things had that little... I don't know, you just had to be there to feel it, it was an amazingly good vibe.

I didn't attend as many talks as I wanted to but, the ones I did, were great: Andy Budd, Delphine Ménard, Fred Oliveira, Julian Bleecker, Mark Wuben, Mike Davis (the real Mike Davis, and he's on Twitter) and Tijmen Schep. All of them brilliant people. I was really looking forward to hear what Julian, Mike and Tijmen had to say and I was not at all disappointed.

Now that SHiFT's over, I'm really, really looking forward to next year's edition.

Just because I'm in the mood for it, I've upgraded the "About me" section of this blog. Blame the weather, the positive thinking I got from shift (more on it tomorrow) or simply the fact that it being so terse as it was came back to haunt me from time to time.

Much is still private but many of the good things are there. If there's anything else you'd like to know email me and I'll add it there.

On this day, I'm Python's top evangelist according to Google:

I was just handed a thought provoking article about the immense discrepancy of living on a big city. Although there are many upsides to that, we're seeing that people want to escape from it but cannot due to lack of infrastructures in less densely populated areas. Focusing on Portugal recent data shows that 42% of the population lives in 5% of the territory. To make matters worse, in 2015 the UN predicts that over 69% of the Portuguese population will live in the two major metropolitan areas: Lisbon and Oporto.

Although I'm not the one with a solution for that problem, this hits home with me because for years I've been advocating the need to create a real technology hub in some rural area of Portugal. This would need to be an artificial conglomerate, much like Silicon Valley.

The TGV (high speed train - or as I like to call it rails broadband haha) is just around the corner and can serve as a catalyst for such an initiative as it not only takes care of the transportation part but it also allows to create something right in the middle of two European capitals: Lisbon and Madrid. Because of that, it should be able to attract people from both countries and from there start a snowball effect, maybe enough to attract people from other countries.

The Portuguese government would of course need to offer some incentives to drag people and companies outside of their comfort zones: lower taxes, maybe even tax exemption on technology products bought on the first years of a startup's life, start a few initiatives for construction works (housing, office spaces and whatnot), infrastructure (fiber to the win!),etc. I'm usually not one to bank on the government to do something but this seems to fall exactly under their area of action.

I'm bubbling with ideas, much more than my poor English skills allow me to write, but what do you think of this? Does it make sense?

On and off I check on Sun's research projects to have a peek into what they're up to. You see, all the good stuff at Sun is classified as research project and they act like Xerox demoing stuff at anyone who cares at PARC until a handful of bright people comes along and starts a revolution... but I digress.

One of the projects I keep tabs on is Project Caroline (how's stupid is that? please name it something more meaningful like Sun Java Horizontal Scaling Services - yes, I'm taking a pot shot). Good thing it's moving forward, ableit in the same way as a glaciar. Here's a good description of what it's all about:

The platform takes care of the details of finding a machine for the process to run on, configuring the machine, network, and Internet connectivity. Operating system-level virtualization is used to isolate processes sharing the same physical machine while keeping per-process overhead low. Customer programs are expressed in languages like Java byte code, perl, and python that provide OS and instruction set independence. Other resources include IP sub-nets, network file systems, databases, external IP addresses, L4 and L7 load balancers, and DNS bindings. Applications can allocate, configure, and release these resources using the platform API. Through the platform API, applications can acquire and release resources in seconds.

How's that for a technical brief? Unfortunately access to the grid is available exclusively to select partners. Time to go try to pull some strings at Sun I guess.

Twitter

Everyone's dog says/blogs/twits that Twitter is a synonymous to fail. They've had more than their share of downtime and keep cutting on features to be able to cope with the load.

Although it's annoying at times, I reckon they're being really, really smart about their options. You see, no matter how bad Twitter performs, no one's leaving. In the meantime, they're tuning their software and systems to make it scale like mad. When it's finished, problems will be a thing of the past, they'll be able to start adding new features and, more importantly, they'll be prepared to cross the dreadful chasm Moore wrote about nearly two decades ago. At that point, it'll be sold to the public as SMS for the Web and, taking a look into where the mobile industry is possibly heading, I can see them making it really big (a big clash of interests will surely follow).

iPhone prices

I keep reading about iPhone 3G prices everyday. I don't mean the price you pay when you get the phone but the sum of everything when the contract finishes. The latest (sorry, portuguese link, scroll to the bottom and ignore the rest) tells the iPhone costs about 1000 Euros after two years.

Man, these people really need to learn a thing or two. The correct way of figuring out how much the iPhone costs is to subtract the value of any contract plan with subsidy from the value of the same plan without subsidy. Example: for the iPhone Best 100 (which is the cheapest), you'll need 299.9 Euros to bring it out of the store. The phone is subsidized and the contract costs 29,9 Euros per month, which isn't a lot if you think about it: 100 minutes of voice on prepaid costs about 25 Euros. If the phone weren't subsidized, the same plan would be 14,9 Eur per month. So to calculate how much the phone will cost you, you subtract 14,9 from 29,9, which gives you 15 Euros and then multiply by the contract duration, which is 24 months, so the result is 360 Euros. Now add it to the price you paid to take it home with you and you'll get 659,9 Euros. This is the real iPhone price, not 1100 Euros, which includes the costs of communications (voice, SMS, data).

iPhone apps

Speaking of the iPhone, a topic came up the other day over lunch with the rest of the 7syntax crew. Someone, who shall remain anonymous :-), was saying that 3rd-party consumer apps will give a big boost to the iPhone popularity. I don't think so, but I can see how people might believe in this given the amount of hype surrounding both the iPhone and it's apps. With that in mind, if I were a betting man, I'd bet against companies who focus their development efforts of iPhone apps for the consumer market. A small note: I'll gladly accept being proven wrong as the game is just getting started. Another note: I'm talking regular apps, not games.

The major barrier to 3rd-party iPhone apps is the language you need to write them on: Objective-C is not the same as C. Where a C++ developer picks up Symbian in no time or a Java developer creates something for J2ME really quick, unless you have some Objective-C under your belt it'll take some time wrap your head around Objective-C and Cocoa. Also, there aren't as many experienced developers available as there are for other platforms. And then there's also the fact that one needs specific hardware (ie. a Mac) to be able to develop the software.

Finally, there's history. After Symbian got popular (ie. when Nokia started churning Symbian phones like mad), some companies focused themselves on developing consumer apps for Symbian. To be honest, I can't name any as still being alive and kicking. The way I see it, the only platform that stood the test of time is J2ME and games contribute a lot to this. What happens is that the general consumer market isn't looking to replace their computer with their mobile phone, so 3rd-party apps that aren't gateways to something bigger are pretty much doomed by default. It may happen, but it'll take lots of time.

Update: both Celso and Pedro made good points on Twitter with regards to the iTunes App Store, which I missed. Still, how is App Store any different form a carrier's own application portal which is also easily accessible on mobile phones? Is it so different and so engaging that it'll prove as important as the iTunes Music Store was for the iPod?

I'm not a tea fanatic. Sure, if the kettle's on I'll have a go but my favorite work beverage is, by far, coffee. Even so, a few days ago a batch of Adagio teas arrived through mail. I placed the order mostly out of curiosity, due to them having so many weird flavored teas available and the samplers costing only $2 USD.

In spite of a minor confusion that caused some shipping delays, the ordering process was simple enough and getting a package from the States delivered in less than a week is a sure way of getting a repeat customer.

These are the teas that were delivered at my doorstep:

• Iced tea delight
• Earl Grey bravo
• Citron green
• White peach
• Rooibos vanilla

Except for the first one, I've already tasted all the others (actually, I'm sipping through a nice cup of rooibos vanilla as I write this) so a small review is in order.

The Earl Grey is a huge disappointment. From what I've learned the bergamot is an orange-y kind of fruit and that was what I expected. Instead I got a perfume flavored tea. Utter crap and put me on a bad mood as it was the first tea I tasted and started to second guess what the others would taste like. Score: 1, just because it's Earl Grey.

Fortunately the other day I woke up way earlier than usual and prepared a cup of the Citron green tea. Now, this is a very good and light tea. It has a very nice lime flavor, just strong enough to be something worth enjoying. I'm giving it a 3 out of 5.

By coincidence the same thing happened this morning. I got up at about 8:15AM and was looking for something light to fill an empty stomach. At such an early time of day I wasn't quite ready to step outside and go to the café so I prepared a cup of peach flavored white tea and I'm really glad I did so as it's really, really good. It's incredibly light and the strong peach after-taste is, I dare say, awesome. This is my favorite out of the four. Score: 4.

As I wrote before, I'm now enjoying the rooibos vanilla. I never had tasted a rooibos tea before, nor a vanilla flavored one. I'm a fan of vanilla ice cream so I had high expectations that were totally met. One of the reviewers mentioned the fact that rooibos tastes like medicine but I can't spot that taste, only the nice vanilla. I'm giving it a 3.5. It's not as good as the peach white but it's definitely better than the citron green.

And so I'm ending this small tea review, but not without mentioning the mint tea Ruben brought to the office some weeks ago, which kickstarted this small tea adventure. The mint tea is awesome to have on those hot summer days and you should have it really hot as it gets the best flavor out of the mint. But don't prepare lots of it as after about an hour it starts to taste really bad.

All in all, Adagio has a very nice thing going on and their expanding their business to Europe. Although I'll probably keep ordering stuff from the States due to the low dollar price, I really hope they can get some traction over this side of the pond and start having a bit more of variety on their European store. Go check them out.

We have a couple of openings over at 7syntax. We're looking for highly motivated individuals that want to take part on something as special as a startup. If any of the following is you, get in touch by dropping us a word via work@7syntax.com.

Python Developer

Key responsabilities:

• design and implement software back-end components
• program in Python and C
• analyse and improve the performance, scalability and stability of Handivi
• colaborate with the rest of the product team
• proactively look for ways to improve the service

Past experience/desired interests:

• having participated in at least one medium to long term project, Open Source or commercial
• solid skills in programming with Python and C on Linux
• experience with Java
• good knowledge of SQL databases, PostgreSQL earns you bonus points
• solid understanding of messaging systems, server architectures and distributed systems
• design patterns and unit testing
• good communication skills and strong command of the English language

Bonus:

• CS degree
• solid experience with the Django framework
• experience with one or more Amazon services: EC2, S3, SQS
• experience with distributed version control systems
• experience with agile methodologies
• able to travel abroad for training sessions and conferences

J2ME Developer

Key responsabilities:

• lead the development of mobile applications
• interact with the design/user-experience team
• provide technology insight for the definition of project architecture with the back-end team
• research other emerging mobile platforms, e.g. iPhone or Google Android
• proactively look for ways to improve Handivi

Past experience/desired interests:

• having participated in at least one medium to long term project, Open Source or commercial
• deep knowledge on Java Mobile Edition, it's advantages and weak spots, including MMAPI
• design patterns and unit testing
• having an interest in scripting languages, like Ruby
• general knowledge on how mobile networks work, particularly data transport
• excellent communication skills and command of the English language

Bonus:

• CS degree
• hands on experience with distributed version control systems
• Linux lover
• knowledge of agile methodologies
• able to travel abroad for training sessions and conferences

The Operator Perspective panel at the Mobile 2.0 Europe conference was a pretty heated up debate. At a point some members of the audience were going ape which, although entertaining as it may be to the outside observer, wasn't doing any service to those who were looking forward to a rational debate.

Still, there were some highlights that deserve a post-event debate, starting with Anastassia Lauterbach of T-Mobile who shed some light with regards to mobile bandwidth consumption:

• Youtube represents about 17% of all the traffic
• peer-to-peer generates 34% of all traffic (I'm not really sure of this number as I failed to jot it down, please feel free to correct me)

She asked if the audience could figure out these numbers in a pure rhetorical exercise, as an excuse for the tight control and lack of data plan innovation on behalf of mobile operators.

While I can understand her point, in my opinion this is something mobile operators brought upon themselves when they came up with the idea of the USB 3G dongle for the masses.

People aren't using P2P on their mobiles, starting with the fact that it'd kill their battery in no time, and Youtube isn't compatible enough with a large enough range of devices that would cause such high data consumption values.

What's generating all that traffic is people with laptops watching Youtube funny videos while they commute or others that plug the dongle onto a computer and want it to be a complete replacement for fixed broadband, which is a completely legitimate assumption as it is what is being marketed by mobile operators.

At the end of the day, data traffic generated by dongles cannot be used as an excuse for the lack of data plan pricing innovation, which is something that was already going on way before such devices became popular.

Another hot topic was that of tight operator control over what gets into the phone (and what it is allowed to do). This is more of a problem in the US than in Europe, as European MNO's are getting increasingly more open although we're not quite there yet. Still, the old excuse of "we are the ones who get support calls" is getting worn out as a recent market survey rates Nokia as the most popular mobile operator and I doubt Nokia gets a flood of support calls, or considers it a problem large enough to become pretty vocal on it, over 3rd-party applications installed on their devices.

Mind you, I don't doubt MNO's call centers get support calls for 3rd-party applications but I also don't believe that by exercising a tight control over what gets installed will reduce them to zero.

A question that was left unanswered was about whether or not the iPhone, and the high data consumptions done by their users, will become MNOs worst nightmare as they have zero control over what gets installed and must "compete" on flat-rate data pricing amongst themselves.

Conclusion

Both these issues, coupled together, prove Max Niederhofer, of Atlas Ventures, right when he says that operators wanting to get a piece of everything makes him, and other VCs, extremely wary of any serious investment in the mobile industry.

When comparing the mobile services/applications ecosystem with what happens on the Web, one has to admit innovation is, in fact, stifled. And this is the reason any open debate between mobile operators and 3rd-party developers will continue to be heated up for the foreseeable future.

So you're planning on ordering a batch of Moo cards and are wondering about what images to use? Are you a photographer, casual or professional, who has a batch of really nice photos on Flickr and likes to see your work being shown to the world? Then, this Flickr group can be of use to you.

It started with Armando's idea of asking people to send him photos to include on his next batch of Moo minicards and this is the natural evolution of it. If you're a artist looking to spread your work, it just doesn't get any better than this. Join the group, share your stuff, we'll take care of the rest. Just a final note though, your work must be licensed under a Creative Commons that allows us to use it. That's it, see you there.

At Mario's I wrote that I too was eager to get my hands on an iPhone 3G, the $199 USD version which is about 125 Euros (of course it won't go out at this price but let's daydream for a while). But I think I pulled the trigger a bit soon because I paid a visit to the online Apple Store only to spot the following, roughly translated, small print:

The iPhone includes up to two years of free technical support, for the duration of the contract with your mobile operator

WHAT?! A two year contract? Unless it's a competitive monthly fee, e.g. 35 Euros, unlimited data (but of the actually unlimited kind), I won't really consider it. At least until someone finds a way of breaking the operator lock so that I can use it with my pay-as-you-go voice call service plus 100MB of mobile data (you know, the small and compact kind) that currently costs me roughly 20 Euros per month[1].

[1] Expect a full explanation of these costs sometime later tonight.