A long, long time ago, I messed up while updating my (Arch) work box,
and ended-up deleting /etc/fstab. Oops. Having to get back to work, I
just hacked the two lines absolutely necessary to boot the box---the
mounts for /home and /---and made a mental note to fix it properly
later. I should have known better.
Predictably enough, I
ended up never doing it---until now. Why? Well, because I use stable,
and for a long time there had been no kernel updates---until now.
Because /boot was unmounted, from the update process' point of view, it
was a regular folder, so all the files that were modified were just
dumped in there. Worse, because the /boot line did not exist in /etc/fstab,
the system had no idea that /boot is where the boot partition is
supposed to be mounted---otherwise during the upgrade process a very
visible warning would have been displayed (I confirmed this specifically
re-testing this scenario). And the day after, I turn on the box, only
to discover that it boots just fine, but that's about it. After booting,
in X, no keyboard, no mouse, no anything---because the system has the wrong modules loaded! (it has the ones that correspond to the kernel in the /boot folder, rather than the /boot partition, which is the one that is actually running, i.e. loaded during the boot process).
And so I had to borrow 5 minutes at a colleague's computer,
to browse for the solution, so I can boot, so I can downgrade the
kernel, fix fstab, re-upgrade the kernel, and finally get back to work.
Murphy, I effing hate you!
Thursday, 10 November 2011
Sunday, 6 November 2011
Say no to ACTA!
I join my voice to that of countless others who fight to, among other things, keep the internet open and accessible to all, instead of a crippled chimera of corporate interests, where you are guilty until proven innocent---and all in the name of profit of a select few, who are already so wealthy that they can afford to sidestep democracy when that suits their purposes.
More information here: http://www.laquadrature.net/ACTA
More information here: http://www.laquadrature.net/ACTA
Sunday, 30 October 2011
On selling, part II
In the previous post, I argued that there's not much point in selling information, the same way you sell books, essentially because you can replicate information ad nauseaum. In this post I wish to make the point that, while it is indefensible to argue that exchange of information between private parties without commercial interest should be regulated, there is a case to made regarding regulations of exchanged information in the context of business.
Before delving deeper into this, I'll just give two examples: I can buy a book, and digitize its information and give to someone else. No law should stop this behaviour. But I should not be allowed to sell that information, for the right to commercialize it should (during some time) be an exclusive of the book's writer. The same could be said about software. While it is unjustifiable to have the legal system meddle in a private individual's communications just to protect economic interests, it is justifiable to do so in order to regulate commercial activity. This was the original scope of copyright law. Of course, back then, there was no easy way to exchange huge amounts of information (whether for commercial purposes or not). But nowadays there is, and so the only way to enforce a monopoly on information trade (which is what copyright essentially is) is to regulate the communication between the parties that are trading it. Which is, like I said above, acceptable for business, but unacceptable for the private communications of the citizenry.
What this means is that you can establish a functioning marketplace, without trumping basic privacy rights. But wait, does that mean that someone might try to corrupt the idea outlined above, and declare that to buy a book, you must first sign a form declaring that you will not share with anyone? After all, buying a book is a commercial transaction, right? Well yes, but unless you're buying it to be used in a (for-profit) enterprise, then the transaction should fall outside the scope of copyright. To put this another way, if you are using someone else's work (or more accurately, the information produced from it) to make a profit, then it is acceptable to have the law regulate that process, and guarantee a fair retribution to that someone else (as long as the copyright has not expired). This incentives innovation, and the creation of new works---exactly the result one expects from copyright.
Remember that we are talking about information that was published (i.e. made available for sale) voluntarily: at no point was there anything resembling information theft. What I propose is that information made available in such a fashion can be used and distributed by private parties with no commercial intent, and no interference from the law. It is only when using such information in profit-seeking enterprises that it becomes justifiable to regulate the process through legislation.
Before delving deeper into this, I'll just give two examples: I can buy a book, and digitize its information and give to someone else. No law should stop this behaviour. But I should not be allowed to sell that information, for the right to commercialize it should (during some time) be an exclusive of the book's writer. The same could be said about software. While it is unjustifiable to have the legal system meddle in a private individual's communications just to protect economic interests, it is justifiable to do so in order to regulate commercial activity. This was the original scope of copyright law. Of course, back then, there was no easy way to exchange huge amounts of information (whether for commercial purposes or not). But nowadays there is, and so the only way to enforce a monopoly on information trade (which is what copyright essentially is) is to regulate the communication between the parties that are trading it. Which is, like I said above, acceptable for business, but unacceptable for the private communications of the citizenry.
What this means is that you can establish a functioning marketplace, without trumping basic privacy rights. But wait, does that mean that someone might try to corrupt the idea outlined above, and declare that to buy a book, you must first sign a form declaring that you will not share with anyone? After all, buying a book is a commercial transaction, right? Well yes, but unless you're buying it to be used in a (for-profit) enterprise, then the transaction should fall outside the scope of copyright. To put this another way, if you are using someone else's work (or more accurately, the information produced from it) to make a profit, then it is acceptable to have the law regulate that process, and guarantee a fair retribution to that someone else (as long as the copyright has not expired). This incentives innovation, and the creation of new works---exactly the result one expects from copyright.
Remember that we are talking about information that was published (i.e. made available for sale) voluntarily: at no point was there anything resembling information theft. What I propose is that information made available in such a fashion can be used and distributed by private parties with no commercial intent, and no interference from the law. It is only when using such information in profit-seeking enterprises that it becomes justifiable to regulate the process through legislation.
Saturday, 22 October 2011
On selling
Selling is exchanging something---anything---for money. Now, there's not much point in selling something that nobody cares to buy. And, from the consumer's ("buyer")
point of view, there's absolutely no point in buying something that you
either don't want, or if you do want it, that you can make it yourself.
In other words, for that exchange to take place, there must be "something in it" for the seller, as well as for the buyer.
This model has stood the test of time: it is as old as business itself. Underlying it, is a fundamental assumption: that after some good has been exchanged for some amount of money, just as the buyer is no longer in possession of that amount of money, the seller is no longer in possession of the good that has been sold. This works well specially when what's being sold either can't be copied, or when doing so would have a higher cost to the buyer than the seller's price. But what happens when what's being sold is something as ethereal as, say, information?
Information might be ethereal, but we, humans, are not. Therefore, with the possible exception of when we're thinking about it, we have to bind information to some material support. Before digital technology, that binding use to be definitive: once you've printed a book, you cannot use the same ink and paper to print a different book. But when someone sells a book, he's not selling just the material support, or just the information on it---he's selling the whole bundle. And while it's true that to the buyer of a book, the information is more valuable than the support it exists on, it is not less true that information without some kind of support is largely useless. Except for the lottery numbers, or the code to arm a nuclear warhead, nobody buys information without some physical support (just imagine what it would be like to sell the contents of The Da Vinci Code, in its platonic ethereal form---a true Mission Impossible). And for this reason, there is not much point in selling information, per se. Or is there?
Enter digital technology, and some years later, the internet. Of the many changes, good and bad, that these two advances in technology brought to society, the one that is clear beyond the shadow of a doubt, is that information is no longer bound to some material support, and irreversibly so. We still need some form of support to interact with information, viz. screens and keyboards, etc. but we no longer have to transmit the support in order to transmit the information. Furthermore, when we say transmit, we actually mean copy: after I send something to you, we both have it. Now it is paramount to understand the magnitude and the implications of such a change. If before the internet age selling a book (or a newspaper) could be considered a way to transmit information (and to fund the work needed to compile and organize that information), afterwords, selling information is, at best, a one time deal. Because that first buyer can then spread that information world wide, with essentially no effort or additional cost. Obviously, that would have been impossible before the internet age. This poses an enormous challenge to the enterprises that base their business models on distributing information when it was permanently bound to some material medium---paper, plastic slabs, magnetic tape, or whatever else. But it does not change the fact it makes no sense to sell information in and on itself. Also note that after that first sale, both the seller and the buyer are now in possession of the information that was sold.
To explain this from another angle: what justifies that first sale is that the information has value to the buyer, and that's why he pays for it to the seller. I.e. he pays for the work needed to produce that information. But nobody will pay the seller for the distribution of copies the information he produced, because the internet has removed all economic value from the process of distribution (because now anyone can do it, hence nobody is going to pay for it). This has always been the case with mathematics and physics: as any book written about those subjects (no matter how valuable to science) is unlikely to become a best seller, funding could not come from distribution. Given that now distribution brings ever more dwindling revenues, such a scenario will become more widespread.
But we're going astray. My contention here is that you cannot sell information in and on itself. And that is not because information is somehow special: you cannot sell anything that the buyer can easily do or copy himself, because, well, he's going to do or copy it himself. There's no added value. In the case of information, there is (a lot) of added value in producing it, and so people will pay for it (yours truly here included), but there is absolutely no value in retail selling it, for the reasons set forth above.
This model has stood the test of time: it is as old as business itself. Underlying it, is a fundamental assumption: that after some good has been exchanged for some amount of money, just as the buyer is no longer in possession of that amount of money, the seller is no longer in possession of the good that has been sold. This works well specially when what's being sold either can't be copied, or when doing so would have a higher cost to the buyer than the seller's price. But what happens when what's being sold is something as ethereal as, say, information?
Information might be ethereal, but we, humans, are not. Therefore, with the possible exception of when we're thinking about it, we have to bind information to some material support. Before digital technology, that binding use to be definitive: once you've printed a book, you cannot use the same ink and paper to print a different book. But when someone sells a book, he's not selling just the material support, or just the information on it---he's selling the whole bundle. And while it's true that to the buyer of a book, the information is more valuable than the support it exists on, it is not less true that information without some kind of support is largely useless. Except for the lottery numbers, or the code to arm a nuclear warhead, nobody buys information without some physical support (just imagine what it would be like to sell the contents of The Da Vinci Code, in its platonic ethereal form---a true Mission Impossible). And for this reason, there is not much point in selling information, per se. Or is there?
Enter digital technology, and some years later, the internet. Of the many changes, good and bad, that these two advances in technology brought to society, the one that is clear beyond the shadow of a doubt, is that information is no longer bound to some material support, and irreversibly so. We still need some form of support to interact with information, viz. screens and keyboards, etc. but we no longer have to transmit the support in order to transmit the information. Furthermore, when we say transmit, we actually mean copy: after I send something to you, we both have it. Now it is paramount to understand the magnitude and the implications of such a change. If before the internet age selling a book (or a newspaper) could be considered a way to transmit information (and to fund the work needed to compile and organize that information), afterwords, selling information is, at best, a one time deal. Because that first buyer can then spread that information world wide, with essentially no effort or additional cost. Obviously, that would have been impossible before the internet age. This poses an enormous challenge to the enterprises that base their business models on distributing information when it was permanently bound to some material medium---paper, plastic slabs, magnetic tape, or whatever else. But it does not change the fact it makes no sense to sell information in and on itself. Also note that after that first sale, both the seller and the buyer are now in possession of the information that was sold.
To explain this from another angle: what justifies that first sale is that the information has value to the buyer, and that's why he pays for it to the seller. I.e. he pays for the work needed to produce that information. But nobody will pay the seller for the distribution of copies the information he produced, because the internet has removed all economic value from the process of distribution (because now anyone can do it, hence nobody is going to pay for it). This has always been the case with mathematics and physics: as any book written about those subjects (no matter how valuable to science) is unlikely to become a best seller, funding could not come from distribution. Given that now distribution brings ever more dwindling revenues, such a scenario will become more widespread.
But we're going astray. My contention here is that you cannot sell information in and on itself. And that is not because information is somehow special: you cannot sell anything that the buyer can easily do or copy himself, because, well, he's going to do or copy it himself. There's no added value. In the case of information, there is (a lot) of added value in producing it, and so people will pay for it (yours truly here included), but there is absolutely no value in retail selling it, for the reasons set forth above.
Sunday, 16 October 2011
&$%¹@£§, parte 2
Quando escrevi há uns dias o que achava do governo e das suas portagens, foi antes de ter ouvido o discurso do Passos Coelho sobre a nova hiper-austeridade (esperemos que o prefixo não fique desactualizado nos próximos meses/anos). O que eu penso disto está razoavelmente bem explicado aqui, pelo que não vou repetir os argumentos. Parvos não somos outros que não nós, que ficamos e assistimos a isto.
De há uns meses a esta parte (desde de que o actual governo tomou posse), tem sido conversa recorrente cada vez que se fala na "crise das dívidas soberanas", realçar o facto de que "Portugal não é a Grécia", não estamos na mesma situação, nem lá vamos chegar. Também espero que não lá cheguemos, mas depois da última quinta-feira, ficámos muito mais parecidos...
De há uns meses a esta parte (desde de que o actual governo tomou posse), tem sido conversa recorrente cada vez que se fala na "crise das dívidas soberanas", realçar o facto de que "Portugal não é a Grécia", não estamos na mesma situação, nem lá vamos chegar. Também espero que não lá cheguemos, mas depois da última quinta-feira, ficámos muito mais parecidos...
Thursday, 13 October 2011
Outra vez a &$%¹@£§ das portagens nas SCUT
Isto já me irrita solenemente. Primeiro o ex-ministro das Obras Públicas dizia que as portagens nas SCUTS era uma questão de justiça e igualdade, tacitamente deixando de fora o facto de que, pelo menos no caso da A23, esta foi construída por cima (literalmente!) tanto do IP2 como do IP6. Entre outras coisas, é este tipo de "meias-verdades" que dá tão mau nome à política e aos políticos. Pois se a A1 foi construída como uma alternativa, pensada desde do início para ser portajada, e construída com o intuito de dar lucro directo, já a A23 foi pensada ser a alternativa, que nunca teria portagens, e portanto a questão de outras alternativas (sem portagens) naturalmente não se pôs. Querer equiparar as duas situações é extremamente desonesto.
E como este, há muitos mais argumentos que justificam o não portajar as SCUT. Aos quais o governo dirá sempre que tem que ser porque não há alternativa. E chegados aqui, como devem os visados por mais este "inexorável" disparate proceder?
Eu começaria por assumir que construir a A23 foi, muito provavelmente um erro. O que eu quero dizer com isto, é que se se aplicar o mesmo critério que se aplicou à A1, a inevitável conclusão é a de que a A23 não deveria ter sido feita. O interior também ficava bem servido com uma IP bem mantida, e com extensas duas faixas, mas sem todos os bells and whistles de uma auto-estrada. Depois, erro ou não, a verdade é que a A23 foi feita, sem custos, e sem alternativas. Se o estado quer agora portajá-la, então que construa primeiro a alternativa, e portaje depois---tal como sucedeu com todas as auto-estradas portajadas à nascença. Cobrar apenas portagens é injustificável. É claro que o estado tem o poder de o fazer na mesma, mas então pelo menos tenha a decência de ser honesto, e diga aos visados, sem margem para dúvida, que o que vão fazer é errado, é punir as populações por um eventual erro que, a ter existido, não lhes é imputável, mas que o vão fazer na mesma porque sempre dá mais uns cobres, e até têm uma pseudo-justificação que, apesar de não justificar coisa alguma, aparenta-o o suficiente para uma grande fatia da mui embrutecida população comer e calar. Ou neste caso, pagar e calar.
E em vez de aumentar receitas, podia-se reduzir a despesa: acabava-se com as estações de serviço e com os postos de SOS, diminui-se a iluminação, etc. Não ficava ao mesmo nível do custo de manter uma IP, mas também evitava isolar (ainda mais) o interior do país, asfixiando ainda mais as economias locais, como inevitavelmente vai acontecer.
Mas o facto de esta intenção do governo de portajar as SCUT não ter justificação, é uma faca de dois gumes. As maiorias democráticas não transformam quem as obtém em déspotas temporários. Quem toma decisões presta contas das mesmas. E ao tomar uma decisão cuja única (ou principal) "justificação" não resiste nem à mais elementar análise, o governo está indirectamente a legitimar formas de protesto muito mais fortes do que simples manifestações ou buzinões ou marchas lentas. Está a legitimar a desobediência civil.
Recusar pagar a portagem que nos querem impor será, muito provavelmente, ilegal. Contudo, aceitar pacificamente uma lei injusta, injustificada, e cujo benefício a médio e longo prazo é no mínimo, dúbio, é aceitar o despotismo do estado. É aceitar que os fins justificam os meios. No limite, é aceitar o fim do estado de direito. E a isso, ninguém deve ficar indiferente.
E como este, há muitos mais argumentos que justificam o não portajar as SCUT. Aos quais o governo dirá sempre que tem que ser porque não há alternativa. E chegados aqui, como devem os visados por mais este "inexorável" disparate proceder?
Eu começaria por assumir que construir a A23 foi, muito provavelmente um erro. O que eu quero dizer com isto, é que se se aplicar o mesmo critério que se aplicou à A1, a inevitável conclusão é a de que a A23 não deveria ter sido feita. O interior também ficava bem servido com uma IP bem mantida, e com extensas duas faixas, mas sem todos os bells and whistles de uma auto-estrada. Depois, erro ou não, a verdade é que a A23 foi feita, sem custos, e sem alternativas. Se o estado quer agora portajá-la, então que construa primeiro a alternativa, e portaje depois---tal como sucedeu com todas as auto-estradas portajadas à nascença. Cobrar apenas portagens é injustificável. É claro que o estado tem o poder de o fazer na mesma, mas então pelo menos tenha a decência de ser honesto, e diga aos visados, sem margem para dúvida, que o que vão fazer é errado, é punir as populações por um eventual erro que, a ter existido, não lhes é imputável, mas que o vão fazer na mesma porque sempre dá mais uns cobres, e até têm uma pseudo-justificação que, apesar de não justificar coisa alguma, aparenta-o o suficiente para uma grande fatia da mui embrutecida população comer e calar. Ou neste caso, pagar e calar.
E em vez de aumentar receitas, podia-se reduzir a despesa: acabava-se com as estações de serviço e com os postos de SOS, diminui-se a iluminação, etc. Não ficava ao mesmo nível do custo de manter uma IP, mas também evitava isolar (ainda mais) o interior do país, asfixiando ainda mais as economias locais, como inevitavelmente vai acontecer.
Mas o facto de esta intenção do governo de portajar as SCUT não ter justificação, é uma faca de dois gumes. As maiorias democráticas não transformam quem as obtém em déspotas temporários. Quem toma decisões presta contas das mesmas. E ao tomar uma decisão cuja única (ou principal) "justificação" não resiste nem à mais elementar análise, o governo está indirectamente a legitimar formas de protesto muito mais fortes do que simples manifestações ou buzinões ou marchas lentas. Está a legitimar a desobediência civil.
Recusar pagar a portagem que nos querem impor será, muito provavelmente, ilegal. Contudo, aceitar pacificamente uma lei injusta, injustificada, e cujo benefício a médio e longo prazo é no mínimo, dúbio, é aceitar o despotismo do estado. É aceitar que os fins justificam os meios. No limite, é aceitar o fim do estado de direito. E a isso, ninguém deve ficar indiferente.
Wednesday, 21 September 2011
Not Google+
An excellent parody of G+, and, indirectly, of all other social networks. Also, if you can read portuguese, this is a very good reading (the video embedded was in one of the comments).
Saturday, 10 September 2011
Emprego, inovação & a "big picture"
Um colunista da CNN diz que os empregos estão obsoletos, o que levou a uma longa discussão no HN. Aqui pelo burgo, onde obsoletos ou não, os empregos são cada vez mais uma raridade, o senhor do vídeo em baixo embebido diz que o que é preciso é que a malta que sai das universidades, em vez de mandar CV's para as empresas, e depois dizer: "pronto está aqui o meu CV, agora arranje-me um emprego" é preciso que o pessoal aprenda a vender o seu trabalho. Toda a gente que eu conheço teve uma reacção unânime ao ver este vídeo: elogios de louvor. Bem, toda a gente menos eu.
Não quero com isto dizer que o senhor esteja errado naquilo que diz: muito pelo contrário, acho que está absolutamente certo. No fundo o que ele diz é que cada trabalhador deve saber ser um "micro-empresário": deve saber "vender" o seu "produto" (neste caso, o seu trabalho). O conselho não é novo: já em 2005, Joel Spolsky dizia essencialmente o mesmo:
Então, qual é a minha objecção a isto?
Simples: isto empurra-nos para uma sociedade que tem uma dependência crítica da inovação. Esta última, que tanto nos deu de bom, quando forçada deste modo, faz com que comece a haver um incentivo para que se resolvam problemas que ninguém tem. Para usar uma expressão mais sucinta, cria-se um incentivo para começar a "manufacturar a procura" (manufacturing demand). Dito de outro modo, a necessidade de inovações que continuem a fazer com que o dinheiro circule é tanta, que começa a fazer sentido que para além de criar um produto, se crie também o mercado para esse produto. O marketing e a publicidade passam a ser tão (ou mais!) importantes do que o produto sobre o qual incidem.
E qual é o problema que isto tem?
Só incentiva as "inovações" que dão lucro, e incentiva muito mais as que o fazem a curto prazo, do que as que dão lucro a longo prazo (as empresas só começam a pensar no longo prazo depois de o curto prazo estar assegurado). E isto é mau por várias razões. Primeiro, porque torna quase impossível que alguém que não seja rico dê algum contributo à sociedade que não seja gerador de lucro---e há muitas formas de riqueza que vão para lá do dinheiro. Segundo, isto leva-nos a um ciclo vicioso em que a dependência da inovação é cada vez maior. Se por alguma razão, houver algum período em que o ritmo a que surgem empresas novas e ramos de negócio novos diminui---e assumindo que os padrões demográficos se mantêm---isso significará mais desemprego, com todas as consequências que daí advêm. E depois lá virão os political pundits do costume, dizer como é necessário investir na inovação, na criação de emprego, no aumento da produtividade, etc. Em suma, continuar a tratar os sintomas, enquanto se faz de conta que a doença não existe.
Miguel - Prós e Contras (20-06-2011) from José Carlos Oliveira on Vimeo.
Não quero com isto dizer que o senhor esteja errado naquilo que diz: muito pelo contrário, acho que está absolutamente certo. No fundo o que ele diz é que cada trabalhador deve saber ser um "micro-empresário": deve saber "vender" o seu "produto" (neste caso, o seu trabalho). O conselho não é novo: já em 2005, Joel Spolsky dizia essencialmente o mesmo:
Why should CS majors learn econ? Because a programmer who understands the fundamentals of business is going to be a more valuable programmer, to a business, than a programmer who doesn't. That's all there is to it.É claro que isto coloca todos os que escolheram outro ramo de estudos que não gestão, marketing ou economia, numa posição um pouco inglória, pois vão acabar por ter que estudar essas matérias na mesma. Mas esse parece ser o caminho a percorrer para criar mais emprego: criar mais empreendedores.
Então, qual é a minha objecção a isto?
Simples: isto empurra-nos para uma sociedade que tem uma dependência crítica da inovação. Esta última, que tanto nos deu de bom, quando forçada deste modo, faz com que comece a haver um incentivo para que se resolvam problemas que ninguém tem. Para usar uma expressão mais sucinta, cria-se um incentivo para começar a "manufacturar a procura" (manufacturing demand). Dito de outro modo, a necessidade de inovações que continuem a fazer com que o dinheiro circule é tanta, que começa a fazer sentido que para além de criar um produto, se crie também o mercado para esse produto. O marketing e a publicidade passam a ser tão (ou mais!) importantes do que o produto sobre o qual incidem.
E qual é o problema que isto tem?
Só incentiva as "inovações" que dão lucro, e incentiva muito mais as que o fazem a curto prazo, do que as que dão lucro a longo prazo (as empresas só começam a pensar no longo prazo depois de o curto prazo estar assegurado). E isto é mau por várias razões. Primeiro, porque torna quase impossível que alguém que não seja rico dê algum contributo à sociedade que não seja gerador de lucro---e há muitas formas de riqueza que vão para lá do dinheiro. Segundo, isto leva-nos a um ciclo vicioso em que a dependência da inovação é cada vez maior. Se por alguma razão, houver algum período em que o ritmo a que surgem empresas novas e ramos de negócio novos diminui---e assumindo que os padrões demográficos se mantêm---isso significará mais desemprego, com todas as consequências que daí advêm. E depois lá virão os political pundits do costume, dizer como é necessário investir na inovação, na criação de emprego, no aumento da produtividade, etc. Em suma, continuar a tratar os sintomas, enquanto se faz de conta que a doença não existe.
Miguel - Prós e Contras (20-06-2011) from José Carlos Oliveira on Vimeo.
Tuesday, 30 August 2011
Why (he) blogs
In mid-2007, when I started blogging over there, I detailed the reason why I was doing it: in essence, I had some free time, and decided to give blogging a try. Never really thought I'd stick to it. But it turns out I did stick to it, and the reason is simple: writing---instead of just "thinking"---forces you to discipline and organize your thought. The reason I'm getting back at this right now, is because I found a great blog post that gives you a great couple of other reasons to blog. Following the one I already mentioned, here's the one that most resonated with me:
Have fun reading.Blogging is a bit like public speaking, and people are scared of it for good reason. You are really putting yourself out there.
If you do it right, you will stay stupid things every now and then. You will piss people off. It's scary at first to hit Publish and push your ideas into the public sphere.
In the process though you will learn how to live with a thick skin and how to confidently state your opinion. You will also come to be more zen about putting yourself out there in general. In my opinion, these skills are essential for good entrepreneur and hard to learn elsewhere.
Monday, 22 August 2011
Clementine
Like amaroK, without the resource-bloat. It has less features than amaroK, but given the features I actually use, that's a plus.
http://www.clementine-player.org/
http://www.clementine-player.org/
Friday, 19 August 2011
TiddlyWiki
It's small, it's powerful, it's awesome. TiddlyWiki (TW) is a one-file personal wiki, that can be used as a notebook, a journal, and (quite possibly) everything in between. It does have a few setbacks though. The biggest one, is that some configurations, like the username, and default tiddlers (TW word for post), are stored in the browser as cookies. Which means that if you don't save cookie, or change browser, those settings will be reverted to their default values---not good. The solution is using persistent options. In my case, I store these options:
It can be customized ad nauseaum. Most of the settings are stored in special tiddlers tagged with systemConfig. This is the case for instance, with CSS (tiddler named StyleSheet, with no tag associated). There are also a bunch of themes, for you to choose from. Have fun!
chkAnimate: trueOne thing that I have *never* been able to do in TW, is installing plugins automatically---it always fails. Which, it turns out, is not big of a deal, because there's a really simple alternative! It doesn't work for some complicated cases, such as SyntaxHighlighter---which, by the way, is a must have plugin if you're a programmer---but for the majority of plugins, it works flawlessly. Another must-have TW plugin, if you write any math at all, is the Mathjax plugin. I installed it via the alternative method I mention above.
txtUserName:
chkSaveBackups: false
chkAutoSave: true
It can be customized ad nauseaum. Most of the settings are stored in special tiddlers tagged with systemConfig. This is the case for instance, with CSS (tiddler named StyleSheet, with no tag associated). There are also a bunch of themes, for you to choose from. Have fun!
Thursday, 4 August 2011
GPU toaster!
So, to make a long story short, I finally decided to try to make my desktop computer jet engine a bit more quiet. Much to my astonishment, I discovered, with the help of the folks at MBit, that the cause of the jet engine noise was my no longer manufactured Asus nVidia Geforce 9500GT. Or more precisely, its smaller-than-a-cookie fan. Seriously. So I flipped my mental insanity bit to logical 1, and decided to turn it off (I did add another larger fan on the lateral wall of my case).
The result? My GPU's idle temperature is now at a nice and cool 77ºC :P Watching a 20 minute anime raised it to 83ºC. Well, at least now I have a pc that's much, MUCH more silent! Anyway, if anyone knows a silent, (preferably fanless) nVidia GPU with enough horsepower to play FullHD movies, feel free to drop a comment ;-)
Note: I'm not a gamer, so basic 3D render and FullHD is about all I require from a GPU.
EDIT TO ADD: the way to discover where the noise is coming from, is to stop each fan, one at a time, and see which one yields the biggest noise reduction. The proper way to stop a fan is to press your finger against its central spindle, until it stops. It's painless, for both the fan and your finger! ;-)
The result? My GPU's idle temperature is now at a nice and cool 77ºC :P Watching a 20 minute anime raised it to 83ºC. Well, at least now I have a pc that's much, MUCH more silent! Anyway, if anyone knows a silent, (preferably fanless) nVidia GPU with enough horsepower to play FullHD movies, feel free to drop a comment ;-)Note: I'm not a gamer, so basic 3D render and FullHD is about all I require from a GPU.
EDIT TO ADD: the way to discover where the noise is coming from, is to stop each fan, one at a time, and see which one yields the biggest noise reduction. The proper way to stop a fan is to press your finger against its central spindle, until it stops. It's painless, for both the fan and your finger! ;-)
Wednesday, 3 August 2011
Yoono!
Is an amazing Firefox extension (among a couple of other things...), that although intended to ease the management of the ever-growing plethora of social networks we are immersed in, has the über-cool feature of allowing you to use a different profile per Firefox tab! Epic win! Checkout the screencast:
Saturday, 30 July 2011
Blogger, blogspot, and comments
As detailed elsewhere, blogger needs to have third-party cookies enabled for comment posting to work properly. But instead of enabling for the whole wide web, you can authorize them only for blogger.com. Reproducing my second comment in the above post in my former blog:
The problem cause is described here, and a simple fix (in Firefox) is to do this: Edit -> Preferences -> Privacy. Select “Accept cookies from sites”, UNCHECK “Accept third-party cookies”, and on “Exceptions” add blogger.com, and click “Allow”. Now you can comment on blogspot.com blogs.
Untested for blogs hosted outside blogspot.
If after the first attempt to comment a red square appears, hit Submit again, and it should work.
Friday, 29 July 2011
A (long) but simple test
Ramona Fricosu is a Colorado woman who is accused of real state fraud. And in order to prove that accusation, the Obama DoJ has just asked a federal judge to order her to enter the password on her laptop, so as to decrypt its contents. The EFF meanwhile pointed out that this runs afoul of the Fifth Amendment:
Decrypting the data on the laptop can be, in and of itself, a testimonial act — revealing control over a computer and the files on it
As computers and encryption didn’t exist in the Eighteen century, one is left with trying to find the most accurate analogy that would have been plausible in those times; Slashdot provides a bunch of them, for both sides of the argument. However, I don’t think that’s how the Fifth Amendment applies to this case. Said amendment provides that (emphasis mine):
No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury, except in cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger; nor shall any person be subject for the same offense to be twice put in jeopardy of life or limb; nor shall be compelled in any criminal case to be a witness against himself, nor be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation.
The immediately obvious way to link this provision to Fricosu’s case is through the bold part, i.e., to concede that forcing her to reveal the password could amount to self-incrimination, if the information on the laptop could be used to convict her. However, the reason to forbid self-incrimination is not to somehow sanction a defendant being able to hide valuable evidence. It’s to avoid having a “confession” tortured or otherwise coerced out of a defendant (a common practice in 16th and 17th century England). However, that’s not what’s at stake here. Let’s assume that if the laptop was not encrypted, there would no risk of self-incrimination. Then, if the laptop is encrypted, but there was a way to be absolutely certain, beyond all reasonable doubt, that the defendant was in possession of the password, she could be compelled to give it away (or, as the prosecution stressed in this case, not to tell them the password, but just to enter it when the laptop prompts for it). For in that case, anything that’s found in laptop would already be there. In other words, this could, in no way, be construed as a “confession” being coerced out of the defendant—it would be similar to the case where no encryption was present.
But herein lies the rub. You cannot be certain the laptop belongs to the defendant, and furthermore, that she knows (or has ever know) the password. She could have forgotten it (I’m a computer engineer, and this happened to me twice, back in my CS undergrad days). Or, more generally, we could be dealing with an innocent defendant, that somehow came to be in the possession of an encrypted laptop, which he does not know how to unencrypt. If the court was to order said defendant to provide the password—an order he would inevitably fail to comply—then an otherwise innocent person would end up being held in contempt of court. This is the situation the fifth is meant to prevent. The US Supreme Court ruled in Ohio v. Reiner that one of the Fifth Amendment’s basic functions is to protect innocent persons who might otherwise be ensnared by ambiguous circumstances. And this is the reason why in this case, the prosecution’s request ought to be denied.
Note: I’m not a US citizen, but I got interested in this matter because 1) it might set an important precedent to those who work in the CS field, 2) it was deceivingly non trivial, and that got me curious.
Subscribe to:
Posts (Atom)