Alcides Fonseca

I’ve been reading tons of reactions to the launch of the latest best-game-ever GTA IV. I liked Penny Arcade’s the most. And real geeks are also impressed by GoogleMaps-powered Liberty City Map. It is a nice mashup indeed.

Today Luís remembered me of Google Drive where you could drive a little yellow car around the world in real streets. Then it hit me: What about joining the two of then, add some extra features, and get a javascript web-based GTA game, with the top-view, just like the original one (which I’m a fan of!).

Someone with a lot of free time? Anyone? Maybe you?

So Twitter is leaving Rails and I bet Rails programmers are not happy with this. Their #1 success application at a large scale is leaving the building. I guess now it’s more difficult for them to persuade businesses to start using Ruby on Rails.

One the other hand, Jaiku is moving in to Google App Engine and I guess they’re using Python. They’re minimizing the risk of scaling issues (something they could have learned form Twitter’s success). I still wonder what’s the key think that makes them working on Jaiku. Is it Android? Or some killer-feature?

Yahoo has lanched Fire Eagle, a way of providing your location to several services. Good news to Python Hackers, Steve Marshall has published his bindings to the Fire Eagle API. Also in the Python world, Simon Willison has done a really cool mashup envolving FireEagle and Wikipedia (along with some Google Stuff) sticking it all inside a Django Project. And he also explains how he made it.

Among the best of the web I found a link to Richard Crowley’s blog. I found there three cool hacks, specially the last two. The first is PownceFS and implementation of your Pownce friends archives as a filesystem. Systems like this and like DropBox, that connect the desktop and the web are stuff that I believe in. Oh, and take a look at TarPipe when it comes out (or earlier at Take Off!).

The other two are some UserScripts to improve you Google Reader experience. I use it as the my major source of information (email being only for private stuff and twitter… you can’t really call it information, can you?) but I archive stuff in my delicious account, so now I have what I’ve been looking for in a while: being able to bookmark posts directly to Delicious inside Google Reader.

The last one, is another script to see that original link for that post but inside GoogleReader, which I find very handy to comment a post, or see some youtube video GReader insists on not parsing.

Let’s start by big G, they’ve released the Google Contacts API. They’re second to Microsoft who have already did the same, in order to avoid situations where you have to give your email password to social networks, so they can import your contacts. A nice step, I just wish applications start to implement both of the APIs right away.

Now Apple, that has released the SDK for iPhone along with some news like exchange support and so. Like many noticed, they are targeting now the Business market with Air and this iPhone update (only possible to Microsoft new interoperability promise) and I guess they will make greater moves into that area in about one year or so. Regarding their store, they are being selfish and they are going to pay that price very soon. If they have hacked dotMac, won’t they be able to hack iPhone, and create a uTunes where you can get cracked software? And I bet a few developers will prefer this one!

Flock 1.1 was released. I have a bittersweet relationship with Flock. It used to be my main browser, but being an programmer I have some needs Flock doesn’t fulfill (not talking about Firebug). However if I only used the browser for social stuff like twittering, facebook, flickr, del.icio.us and have a wordpress blog, it would be perfect, just perfect.

With the release of Silverlight2b1, a few apps from partners are being shown up, and I really liked TextGlow, a online viewer of docx documents. That made me thing. Currently Office Online doesn’t have a collaborative tool. That makes ALL the difference when comparing to Google Apps. I believe the reason they’ve not released it yet is because they’re getting something really powerful built on Silverlight2. I can’t wait to try it out!

A few months ago I moved to GMail and Google Reader for my main information needs: email and RSS feeds. I now have 3 computers (windows desktop, macbook and HTC TyTN) in which I sync Outlook, Mail.app and WM email client through Gmail’s IMAP. And in Gmail I have my @gmail.com account, me@alcidesfonseca.com and my student email account. So Gmail works both as a aggregator for my emails and as a web interface when I’m not in my computers or in restricted WiFi networks.

However I’m not satisfied with my current situation. Seems like Gmail has an intelligent algorithm for setting the frequency for fetching you POP3 external accounts based on your email frequency. I find myself going to settings windows and force POP3 fetching manually. But this is not pratical! Sometimes I even go to the webmail version of the original email account when I know I should have received an email.

Of course I could set up the POP3 accounts in my clients, but it’s not the same since I need syncing between them, and I like Gmail’s interface (and that’s whats keeping so many people addicted to it). And I could make my other emails redirect to my gmail account, but after some stories about “lost” emails in gmail a year ago, I like to keep my emails in other servers.

I guess they just fit their computing power to the users needs. If I get one email per day, they’re fetching it one or two times per day. If I get one per minute, it fetches every 5 minutes or so. In the beginning I had a five minute frequency that I consider normal, but in the last days, I get “Last checked: 50 minutes ago.” messages. I can’t get a email half an hour after it was sent to me! I should be able to define my maximum delay for emails. Even if I got one email per day, it might be urgent and I need it 5 minutes after it was sent!

And I’m not the only one complaining about this. Does anyone have a script that simulates the manual fetch? It would be really useful!

Brad has announced Google’s Social Graph API. It is indeed cool, but not that amazing. There are a lot of FOAF and XFN parsers around, but being able to poll the web cache they have for google search. This really gets your requests time down to one (the API). Of course, now you don’t get real-time parsing. I really hope Google will work with Technorati’s Kitchen Ping Service but it’s more probable that they launch their own microformat search.

So what’s the feature I miss so much? Contact merger. When I ask for my friends I get this huge list of all my contacts in all my social networks. But I have friends that are in more than one social network, and then being repeated there. It’s possible to merge them, but I would require the maximum of N-1 queries extra from the API. This will really slow it down for friendly people like me. Can’t powerful Google support this in its API?

The first one is not a tech one, but still geeky. Randall Munroe is the author of xkcd, a fantastic comic about romance, sarcasm, math, and language.

The second one is indeed technical. Håkon Lie and Michael Day present PrinceXML and CSS design for printing. A really cool presentation to learn a bit how to style you webpages to generate an handsome print layout or PDF file.

First, Yahoo (and Flickr) will be OpenID 2.0 providers starting 30th this month. Google announced that Blogger is now working a provider but only if you allow it. I really don’t know why they are asking people to check that damn checkboxes, maybe because of their enourmous database. There is not a list of trusts you can manage, for now you just can add one. At least they are already consuming openID. I wished I would not need a yahoo account to have my own space at Flickr. But that’s not for this moment yet.

Microsoft is getting late here. Wouldn’t they like to have the url id.live.com/alcidesfonseca spread all over the internet?

A Segunda Grande Maravilha da Internet faz hoje 7 anos. Quem mo lembra é o Ciberesfera. Mas uma das Grandes Maravilhas da Internet porquê? Bem, eu considero que a Internet está ainda nos seus primórdios e até ao momento há dois marcos bastante importantes na sua história. Os critérios que aplico são a sua influência global, e não apenas nos hackers e novos geeks.

A Primeira Grande Maravilha da Internet foi (e continua a ser) o Google. Refiro-me aqui ao Google como motor de pesquisa e não como empresa nem a todos os outros produtos que tem. É verdade que a sua fortuna se baseia na publicidade online, sobretudo com a introdução de publicidade contextualizada. mas para o utilizador final a marca google está e estará sempre associada à página de entrada do browser. Ou à barra no canto do browser. Ou a qualquer caixinha onde possam colocar umas palavras e encontrar sites relacionados com essas mesmas palavras. Numa altura em que a Internet se estava a tornar demasiado grande para os simples directórios (Como o SAPO em tempos o foi!) e o Google apareceu como uma página que nos levava ao que quiséssemos encontrar. A interface minimalista e a linha mágica de Javascript (e um bom marketing por trás, digo eu) deram ao Google a preferência dos utilizadores e hoje em dia é sem dúvida o motor de busca mais usado. Revolucionou uma geração: a minha. Quando a Internet chegou a preços decentes à minha terriola (a.k.a. 56kbps) todos os nossos trabalhos de pesquisa foram feitos graças a essa página do Google. Belos eram os velhos tempos em que podíamos colocar http://www.google.com na bibliografia! Depois as outras gerações começaram a aprender. Muitos negócios fazem-se através da Internet e o primeiro passo para essa globalização é sempre dado a partir de um dos vários domínios do Google. Hoje em dia está tão integrado na nossa cultura que se tornou um verbo em várias línguas.

A Segunda Grande Maravilha da Internet, a tal que faz hoje 7 anos, é a Wikipédia. Tal como o Google, é um dos sites mais usados pela população em geral. De momento substitui-me as chamadas “bíblias” das cadeiras do meu curso. É usada diariamente por uma data de estudantes como ferramenta de ensino, e por outro tipo de estudantes, que apesar de não andarem num curso, continuam a aprender a vida toda. Qualquer curiosidade é esclarecida prontamente e em várias línguas pela Wikipédia. E há um factor chave aqui: a qualidade. Apesar de ser um projecto que se baseia na boa vontade dos participantes, conseguiu atingir um nível que acaba por ter uma qualidade superior a enciclopédias tradicionais. E muito, muito mais artigos.De facto considero um marco na Rede não só pelo que atingiu, mas para mostrar o produto de uma sociedade muito vasta. O poder da Partilha. E com um resultado brilhante! Com a qualidade, vieram os links da Internet para os diversos artigos (como os que se encontram neste post), e hoje tem quase sempre os primeiros resultados no Google. E agora todos usam.

Aposto que muitos de vós se interrogam porque não considero as Redes Sociais a Terceira Maravilha. Não digo que não o serão, mas de momento ainda não as considero como tal. Para além de existirem diversas redes sociais, e cada uma é mais usada em alguns sítios do que outros ( Hi5 em Portugal, Orkut no Brasil, Facebook e MySpace nos EUA, etc..), ainda não está bastante divulgada nos diferentes extractos etários. São sobretudo usados por jovens, e enquanto não chegarem massivamente ao resto da família, não lhes vou dar essa importância. No entanto estão sim a caminhar para lá. E já o facto de saber de namoros que acabaram por causa de perfis do hi5…. Quando as redes sociais estabilizarem daqui a um ou mais anos, voltamos a falar nisso.

Para já, parabéns Wikipédia!

Ontem uma amiga minha pediu-me para lhe arranjar umas músicas para uma apresentação natalícia que queria fazer para a família(daquelas com muitas animações e com muitas cores que as meninas sabem fazer…) e eu lá usei as minhas capacidades mágicas (também conhecidas como informáticas) para desencantá-las. Antigamente toda a gente recorria ao eMule ou aos torrents, mas hoje quem domina é o HTTP.

Quase todas as músicas consegui usando o motor secreto do Google para músicas. Não conhecem? Então graças à maravilha do opensearch verifiquem uma seta junto àquela barra de procura do Google no topo superior direito do browser (no Flock é um icon laranja à esquerda do endereço) ou simplesmente cliquem aqui. Adicionem o motor de busca “MP3 search” e agora podem procurar músicas facilmente a partir daquela barrinha! Catita, hein?

Mas havia uma que teimava em não ser encontrada. Caviar do Zeca Pagodinho para os mais curiosos. Mas se tenho o vídeo no youtube, bora por a tocar, ligar um cabo da saída de áudio para o line-in, e toca a gravar! Isto era o que eu fazia há uns anitos, mas hoje seria too much hassle, por isso após umas buscas encontrei uma ferramenta muito nice que me convertia vídeos do youtube para mp3, tudo online. Fiquei todo satisfeito!

Mas melhor, para além do iutubiu aceita também vídeos do metacafé, myspace e google videos ou ainda de ficheiros do vosso computador. Quanto a formatos, exporta para mp3, mas também para outros formatos de vídeo como 3GP (telemóveis), mp4, mpg, mov, avi, Flash (swf e flv), avi com DivX ou Xvid e ainda mp4 para iPod ou para PSP. Um canivete suíço completo para riparem conteúdos do Youtube, que por acaso tem uma base de dados de músicas (e de concertos ao vivo!) brutal! A começar por aqui.

(Atenção, eu não vos encorajo a fazer pirataria! Verifiquem primeiro se o conteúdo que pretendem não tem copyrights! Ou então tapem bem as vossas pegadas ;)

Going back a few weeks ago, in a middle of a “Alcides, help me on this computer-related matter” conversion with a non-geek friend:

To which I replied:

That’s it! Big G is number one on the internet. Thanks to Firefox (and other tools), Google is the default page for a lot of folks out there! And number one engine search that I believe is more used than the address bar itself (more on this in the near future)! But how many searches you did in the past week had a Wikipedia page as the first or second result? A lot I bet!

Envy is not a good thing, but people with money need to multiply it and do whatever it takes to expand themselves into what consumers want. Back in 95, when I first used the internet, it was for searching stuff to my schoolworks. It’s still its main purpose for many people. So Google has to own that domain, and Wikipedia has gone too far already! So, let’s spend $1.000.000.(put as many zeros as you want).000 and get professionals to do better content! Users will prefer our database! And luckily they will click on our adsenses that will make us earn $9.999.999.(put as many nines as you want).999!

Smart move, but will the people that already are used to free and without ads wikipedia change to something that provides more quality and has some advertising? I guess no. And most wikipedia articles are already written by guys who know the subject!

It’s sad that Google appears to be that WeSupportOpenSource guys, but when it comes to opensource content they are not that open!

Oh, and this time I’m proud of being Portuguese. Our little google, SAPO has already done something like this. Too bad I don’t think any of them will be useful…

More on this here and here.

OpenSocial is the new fashion among Social Networks. Well, Sérgio (who I will be working with in the next few months) decided to make both a statistics tool for you OpenSocial Widget (or whatever you call it) and also a directory of them all in one.

I do really like that a universal widget system is built, but I was kinda hoping it was based on NetVibes’ UWA that were even compatible with iGoogle… But looks like the Big Brother needs to own everything.

And I already see this OpenSocial widgets running on the Android coming up!

Two of the Web2.0 buzzwords are Accessibility and Usability. They both should be implemented in the websites of this new age of web. Now another buzzword, this one technical: AJAX. This technique to update only some parts of the website without reloading it all has allowed developers to increase the usability of their websites making them faster and improving the user interface to make them resemble desktop applications.

But at what cost? Most of this websites are not correctly loaded on mobile phones or javascriptless browsers. But there are a few that do the AJAX degradation well or pick the easy way (my favorite) and create a basic HTML version. I believe this last one is more effective because the user interface using Javascript with AJAX has a different structure than simple HTML and they should be made independently.

But for those who don’t like making another full website for the minority of browsers, there is now a framework for injecting “accessibility using AJAX techniques”. It’s called AxsJAX, it is being opensourced by Google (It is being used on Google Reader) and you can find out more about this on ifacethoughts.

Usually I do not post about something that everyone is talking about, I usually wait until things settle down to make my point about it, but this time I couldn't resist.

First thing I noticed and you too probably, is that the Android, Google’s OpenSource Platform, offers the GUI, browser and maps pretty much like the iPhone. They probably have planned it, so that the AvantGarde of the Web2.0 and Gadgeteers spend their money on the iPhone, and months later the so rumored gPhone comes out with a look-alike OS, but running on every device. Of course it lacks multi-touch and Apple’s design, but you have the freedom of choosing any operator you like and also it’s an open platform.

Apple lost my (and for sure other’s) respect when they closed the iPhone for 3rd party applications. They basically killed the change for the iPhone to be REALLY used. Of course there is a SDK coming up on February (until then, I will be making my own applications on my Windows Mobile and now, Android…) and I won’t be surprised if it comes out that the iPhone runs Android applications (as they both are UNIX based (iPhone runs on MacOS and Android on Linux) and Android has support for a accelerometer… Oh, and both have a webkit based browser…

Google has made the right choice launching it: first to the developers so they are part of the final product and giving away $10M for the best ones! Nokia had a contest on this and Microsoft runs several small competitions for mobile applications and Apple… hmmm, nevermind. Although we don’t have real devices yet (like OpenMoko is selling), you have a pretty fine emulator (yes, I have tested it) that gives you a perfect idea of what it is with a real device. I look forward to test this and have it running on my Windows Mobile. Of course a lot of apps are lacking and I don’t know what will be shipping the official Google version, since they have it all in the web!

The 3 carrier has announced the SkypePhone, a 3G mobile phone that comes already with a integrated skype client. And calls between Skype users are free! And it really means free since “3 won’t charge you for the data access for VOIP calls”! You can read more on this on Dale Lane’s blog.

I know you can already have skype on your phone, but for most of end-users it is simpler to buy a phone where it already comes installed and configured.

After the removal of VOIP service on the N95, I believe operators are taking the right path towards the ubiquitous web. They shouldn’t block alternatives like VOIP, they should encourage them and increase the value of their natural services.

I hope to see Skype launching the phone all around Europe including Portugal where I know a lot of people that would buy it the first minute it comes out. And seems like the battle for the mobile market has begun. First the iPhone, now the SkypePhone and in the next two weeks Google Phone!

Web applications are borning everyday and in somehow the process to launch them seems very similar, so from what I’ve seen, I’ve written my plans for a generic web2.0 application:

• The Idea - Yeah, I know it sucks, but you'll need that. Really, you cant skip this step!
• The Blog - Before you start to do anything, start a blog where you post any idea. The first posts will have no comments, but with a little SEO magic, you'll get there. Don't forget to post updates frequently.
• The development - This is the boring part of the task, but people say that making it with a small team makes it kinda fun. If I were you I'd use Django or Rails because they come with some buzz. Ok, and they are pretty awesome frameworks too.
• Microformats - Make sure people can use your content in other ways than predicted. Don't forget to include XFN (portable social networks) since all web2.0 applications have a social network included.
• OpenID - Yes, this is the future of "Login/Logout/Sign up" and means less stuff to handle on your side, and will make users happy. If you are targeting a non-tech public, you should enable traditional authentication and make it work as a OpenID provider.
• OpenSearch - Many forget this, but I find it very important, since it integrates your site search with browsers (for example). It will help users to remember them can search for stuff in your application
• Public API - Provide everyone access to their data through Webservices or REST. It will also allow developers to do their mashups and desktop applications that will give you more buzz! And happiness to users!
• Design - KISS. Soft gradients and round corners. Personally I dont like corners too round. Oh, and use famfamfam icons :P
• Go Beta! - Doesn't matter if your product is ready, you should launch in beta. In one way, it is kind of cool to have that badge, but in another you can have bugs or even scaling. It's a good policy. Have in mind using private invites, so you don't have a boom of accesses.
• Talk to TechCrunch - You will need to buy better hosting soon, so you talk to TechCrunch to post about you and from them on everyone will use your website.
• Sell to Google - Google will approach you and you will sell it. You will then live happy ever after in the Bahamas (or not!).

That’s it! Pretty simple, hein? Missed something? Go ahead and comment about it :)

Hoje na aula de Sistemas Operativos, o meu professor abordou o facto de quando acabarmos o curso (Engenharia Informática) nos depararmos com uma feroz competição a nível global. Cada vez mais entram no mercados licenciados indianos e chineses que são 10 vezes mais baratos que um português, por exemplo. Temos então de nos diferenciar pela qualidade. Uma empresa pode ir ao Rent a Coder pedir um trabalho a um indiano que o faz ao preço da chuva. Mas se quiser qualidade vem aqui ao português e paga bom preço.

Ora eu a semana passada ouvi exactamente a mesma coisa pelo Vítor Santos, no lançamento do programa MSP, quando apresentava o concurso ImagineCUP e dizia que era uma óptima oportunidade para mostrarmos ao mundo que somos mesmo bons no que fazemos. Fazendo a comparação com a Teoria dos Búlgaros (em que um ganhou, mesmo não sendo da Bulgária e ficaram conhecidos todos como sendo bons programadores) que precisamos de nos destacar nestes concursos para darmos alguma reputação ao nosso país. Mesmo quem não gosta da Microsoft ou deste senhor em particular tem de admitir que é verdade. De facto este concurso exige que sejam usadas, para além da imaginação que é fundamental, tecnologias Microsoft (que eu até gosto bastante), mas para aqueles que não sentirem assim tanta afinidade existem outros concurso a nível global que poderão apostar, como o Google CodeJam, o TopCoder, ACM-ICPC ou o IEEE CSIDC.

Se acham que fazem programas de topo, não perdem nada em concorrer. Para não falar nos fantásticos prémios que oferecem!

A small rant: No, this is not yet another post saying that Google just bought Jaiku and asking why haven’t they bought Twitter. I’ve seen hundreds of posts like this in the last day and only one or two really commented it. Maybe I should refilter my subscription list.

Now to the point: Google is taking over our world step by step. First searching, then advertising, e-mail and now news-reading and who knows what can be next?
Today while reading my feeds, I saw this mockup that really impressed me:

This is what I have requested! Just a simple OS (yes, Linux based will be O.K.) not for mobile (and why not too?), but for desktop where it could integrate all it’s services. Is it hard to do? Well Google stocks crossed $600 so money is not their problem. They have 13,786 full-time employees and a lot of them are genius. So why are they not launching fantastic applications? I mean, they do, but that’s not as much as I would expect from some big company like them. Things are not that hard, there are user demands for new stuff, so why is Google waiting for?

Any thoughts on this?

As mentioned before, I started to use Google Reader and Mail a while ago and now I have a few ideias for Google to implement:

First of all, why not a optional external CSS file for all my account of google. It would affect my google reader, mail, iGoogle, etc… This matters to power users who chose (as I did) google apps because they access it from various devices. And it wouldn’t cost a thing! Yeah, I use GreaseMonkey but when I use computers that are not mine, it sucks. I don’t like google’s design, it’s too white and blue…

I use Google Talk (in which webcam support is missed a lot) and I love the notifications for mail, but I would like to delete and mark as read right there, because only from reading the title, most times I already know what I should do with it.

Other of the things that I do really miss is a GReader and GMail desktop. Everything would be smoother and much faster running from the OS. I think Google is too ahead of real users, those who aren’t ready to move everything into the browser. In this aspect Microsoft is being slower keeping desktop apps like Windows Live Mail and Writer.

Just some ideas from someone who has been using Google apps almost every hour.

Before putting my hands in the ejaki project, the previous version of the website was pretty old-fashion but used Google Maps API. Wow I though, but then I saw it was only used to get thumbnails for each ejaki (Point of Interest). Well, in my opinion, the map should be the center of the website, just like it is on the mobile app, since Geography is the background of our shared information.

First I had to evaluate the existing APIs: Google Maps was indeed good, proved by the fact it is the most used, and includes features like GeoCoder, remote GML overlays, traffic and driving directions. This should be useful features to add in the future for our application.

Yahoo Maps AJAX (Flash was not an option) was pretty good, with almost all the features as Google’s but with integration with Y! Pipes and GeoRSS. This extras were not relevant since we intend to use only our database (in which other information can be added).

Microsoft’s Virtual Earth shown to be a really nice platform. Microsoft is known by being left behind in the web2.0 age, but this one is from far the most powerful engine. Since 3d visualization, “Birds eye”, powerful information finder, and all the features from both Yahoo’s and Google’s APIs. But there as a downside that matters. Since this is a scientific research project, we don’t like to use proprietary formats, and Microsoft is known for that. We wouldn’t like the possibility of in the future their API would only work on IE.

Open Layers is a free and opensource library, a project of the Open Source Geospatial Foundation. As an opensource alternative it would be nice to use and foment it, but it’s too poor for now. No Satellite view, no geocoding and nothing that would make it more useful for our users. Maybe in the future in another project…

As so, we sticked to Google Maps since it had the basics and some extras that might be useful in the future. Later working with it, I found out it was really nice, but one function was missing: To set the zoom depending on the geocoding answer, and so here you have it for free ;)

In the last days I’ve been using Outlook 2007 because it is the only decent app that syncs perfectly with my PocketPC. In my opinion, it is kind of heavy, I dont like the interface pretty much, and I dont have POP3 sync with the PDA. I also wanted a web-based version to see in the lab computer. Then I decided I would change to Google Mail, Google Calendar and Google Reader because they had a web mobile interface that worked pretty fine in PocketIE. Said goodbye to Flock (we spend wonderful time together…) and installed Firefox 2.0 and the basic addons.

First thing I noticed in Google Mail and Reader was that “Google design” I really hate. Solution: Installed Stylish and then some skins to make it usable to me. Then I noticed Greasemonkey also had the same skins, so I changed the plugin, and now I have the theming + some hacks in one pluggin.

Google is the number one big company when coming to Web Apps, but as they did with Google Desktop, GTalk or GMail mobile ( btw JAVA, so it sucks on Windows Mobile), they should get a desktop application where we had Calendar, Reader, Mail, and maybe others, all together and syncing with the mobile versions( and no, I can’t use POP3 with most of clients, because I have a few other accounts in gmail, and it answers always through the default address).

I am working on a web-based application that will have all the things I use in my department (webmail, sftp, calendar, news, etc) all together, and then I will abandon google stuff, that I dont like using much :/

EDIT: I have tried pyRfeed and after a few hacking, it worked. Not very nice now, but I'll keep an eye on it in the future.