Colpo grosso a Singapore: trafugate 1,5 milioni di cartelle cliniche digitali, inclusa quella del Primo Ministro - il mio articolo per La Stampa

http://www.lastampa.it/2018/07/21/tecnologia/colpo-grosso-a-singapore-trafugate-milioni-di-cartelle-cliniche-digitali-inclusa-quella-del-primo-ministro-3aY5AKqlfOunvI51log6ZM/pagina.html

Il computer pensa con la memoria: ecco cosa è il memory-driven computing - il mio articolo per Il Sole 24 ORE

http://www.ilsole24ore.com/art/notizie/2018-07-20/il-computer-pensa-la-memoria-ecco-cosa-e-memory-driven-computing-091235.shtml?uuid=AEEZHULF

LUn problemino con la firma digitale PGP vecchio di un paio di decenni. Oh my!

Money quote: “Digital signatures are used to prove the source of an encrypted message, data backup, or software update. Typically, the source must use a private encryption key to cause an application to show that a message or file is signed. But a series of vulnerabilities dubbed SigSpoof makes it possible in certain cases for attackers to fake signatures with nothing more than someone’s public key or key ID, both of which are often published online”

https://arstechnica.com/information-technology/2018/06/decades-old-pgp-bug-allowed-hackers-to-spoof-just-about-anyones-signature/

L’articolo su Medium è stato scritto da uno che, più che il silenzio, dovrebbe cercare di imparare uno stile decente di scrittura. Ma, a parte questo, un fondamento di verità c’è ed è interessante. Il silenzio serve.

Money quote: “Mental noise is hurting our minds — we are continually asking questions that create busyness, not knowledge. We are in ‘reacting mode,’ leaving no room for reflection. To regain perspective in life, you need to pause. Silence is fertile ground.”

https://medium.com/personal-growth/why-silence-is-the-think-tank-of-the-soul-9a79049824ca

Una rilettura in chiave kantiana del desiderio sessuale e della conseguente trasformazione in oggetto della persona desiderata, cosa eticamente non bella. Le conseguenze sono surreali. Abbiamo sbagliato tutti tutto?

Money quote: “Is it possible to have sex without objectification? Of course. Prostitutes do it all the time. So do many long-term couples. They have sex with people whom they do not desire. And with no desire, there is no objectification. Not even love can fix it. When the desire is high, when the sexual act is in full swing, my beloved is a piece of flesh. (Though love does lead to occasional cuddling, which is nice.)”

https://aeon.co/ideas/why-sexual-desire-is-objectifying-and-hence-morally-wrong

In ode of GoLang. Un po’ tecnico ma interessante e approciabile.

Money quote: “What are Threads?
A thread is just a sequence of instructions that can be executed independently by a processor. Threads are lighter than the process and so you can spawn a lot of them.“

https://codeburst.io/why-goroutines-are-not-lightweight-threads-7c460c1f155f

Specializzato come la macchina per fare i toast o generalista come i droidi di Guerre Stellari? La risposta è la prima e, quando lo avrete capito bene, il passaggio successivo diventa come trasformare un insieme di tostapane in un sistema complesso e con un dominio di intelligenza più ampio.

Money quote: “Developers who train machine-learning algorithms have found that it often makes sense to build toasters rather than wonder-boxes. That might seem counterintuitive, because the AIs of Western science fiction tend to resemble C-3PO in Star Wars or WALL-E in the eponymous film – examples of artificial general intelligence (AGI), automata that can interact with the world like a human, and handle many different tasks. But many companies are invisibly – and successfully – using machine learning to achieve much more limited goals. One algorithm might be a chatbot handling a limited range of basic customer questions about their phone bill. Another might make predictions about what a customer is calling to discuss, displaying these predictions for the human representative who answers the phone. These are examples of artificial narrow intelligence (ANI) – restricted to very narrow functions. On the other hand, Facebook recently retired its ‘M’ chatbot, which never succeeded in its goal of handling hotel reservations, booking theatre “

https://aeon.co/ideas/the-ai-revolution-will-be-led-by-toasters-not-droids

Le intelligenze artificiali sono diventate uno dei grandi temi popolari del nostro momento storico. Con una connotazione spesso negativa comunque quasi messianica. Con la AI ci fai tutto: se non oggi, almeno domani.

In questo articolo il concetto viene portato avanti un altro po’ e si dice - senza neanche troppo imbarazzo - che la AI potrebbe essere la fine della teoria, perché ci penseranno loro a formulare teorie partendo dai set di dati.

Non è vero, però è subdolo.

Money quote: “We can now assume that given the right datasets, modern AI classifiers can generate high-performing and predictive models without any dependency on the underlying principles that govern a problem.”

https://worldpositive.com/the-new-intelligence-e3e1ff697f11

Ecco a voi (forse) gli iPhone di settembre. Almeno, i loro codici.

Money quote: “Spotted by Consomac, Apple has chosen to publicly file identifiers for all its new iPhones in the Eurasian database, and it confirms three distinct designs will be coming to market.”

https://www.forbes.com/sites/gordonkelly/2018/07/19/apple-iphone-x2-x-plus-se2-cheap-iphone-upgrade-specs-display-release-date-ios12/#32e6377a4d98

Scrivere è solo il principio. Poi viene l'editing. E questa è veramente la parte più difficile. Perché ci si arena, non si va avanti, e si blocca tutto. A parte Raymond Carver, praticamente tutti i migliori scrittorio sono soprattutto dei fantastici editor di se stessi.

Money quote: "This article focuses on editing nonfiction, but I follow a similar process for editing both my fiction and nonfiction writing. You can adapt this advice according to your own writing projects. Read on for my editing tips."

https://writingcooperative.com/how-to-effectively-edit-your-writing-in-7-easy-steps-d24c9468bb57

C’è stato un momento unico e irripetibile nella storia. Quel momento non tornerà più, però sappiate che un tempo i giornali non avevano paura della matematica. E neanche della fisica...

Non so se farà strada, ma l'idea dietro AsteroidOS è interessante: sistema operativo open source per orologi smart. In pratica, permette di sostituire il sistema operativo di molti watch Android e Android Wear. Interessante.

Money quote: "We believe that when it comes to wearable devices, users should have full control over their machines and data. AsteroidOS has been created from the ground-up with modularity and freedom in mind. For this reason, it is a free and open-source project."

https://asteroidos.org

Un indicatore pressoché perfetto della rapidità con cui i sistemi di intelligenza artificiale stanno crescendo è il consumo delle risorse di calcolo (usate per addestrare gli algoritmi, ad esempio). Studio molto interessante.

Money quote: "We’re releasing an analysis showing that since 2012, the amount of compute used in the largest AI training runs has been increasing exponentially with a 3.5 month-doubling time (by comparison, Moore’s Law had an 18-month doubling period). Since 2012, this metric has grown by more than 300,000x (an 18-month doubling period would yield only a 12x increase). Improvements in compute have been a key component of AI progress, so as long as this trend continues, it’s worth preparing for the implications of systems far outside today’s capabilities."

https://blog.openai.com/ai-and-compute/

La categoria-ombrello dell'intelligenza artificiale contiene tante cose diverse. In questo caso, il machine learning e le reti neurali. Una spiegazione ragionata ed esaustiva su come si proceda ad analizzare e risolvere una classe di problemi, in questo caso la classificazione di testi non strutturati (tweet, post su Facebook, email e altro) utili per delle cause legali.

Money quote: "Our goal was to address these two problems: a) deal with NLP problems where we don’t have masses of data and computational resources, and b) make NLP classification easier. As it turned out, we (Jeremy and Sebastian) had both been working on the exact field that would solve this: transfer learning. Transfer learning refers to the use of a model that has been trained to solve one problem (such as classifying images from Imagenet) as the basis to solve some other somewhat similar problem. One common way to do this is by fine-tuning the original model (such as classifying CT scans into cancerous or not—an application of transfer learning that Jeremy developed when he founded Enlitic). Because the fine-tuned model doesn’t have to learn from scratch, it can generally reach higher accuracy with much less data and computation time than models that don’t use transfer learning.

Very simple transfer learning using just a single layer of weights (known as embeddings) has been extremely popular for some years, such as the word2vec embeddings from Google. However, full neural networks in practice contain many layers, so only using transfer learning for a single layer was clearly just scratching the surface of what’s possible."

http://nlp.fast.ai/classification/2018/05/15/introducting-ulmfit.html

A New York ci sono le aste all’incanto di licenze per taxi andati in fallimento causa Uber.

Money quote: “A record 139 taxi medallions will be offered for sale in bankruptcy auction this month — the latest sign that a deluge of ride-sharing apps like Uber are squeezing cabbies out of business and deeper into debt, as well as pinching the incomes of for-hire drivers, according to analysts.

The medallions will be auctioned for a fraction of their original value — some likely having cost their owners as much as $1 million or more apiece”

https://nypost.com/2018/06/09/139-taxi-medallions-will-be-offered-at-bankruptcy-auction/

Un paio di cose da sapere sul ritmo circadiano. Non solo per chi viaggia nei fusi orari, ma per chi sta dentro un tempo sfalsato.

Money quote: “People who are used to having an evening coffee and then have a lot of bright light in the evening and more likely to sleep very late. And since they do it on a daily basis, they think it’s natural to them.

In the book, I talk about the example of my dear friend who is also a circadian biologist. He took his lab on a camping trip. Almost everybody in the lab thought they were night owls because they all go to bed after midnight. But when he took them camping and they had very little access to bright light, all of them became early birds. They went to bed three to four hours earlier and woke up very fresh, early in the morning. So some of the variance is definitely due to external habits.”

https://www.theverge.com/2018/6/12/17453398/sleep-circadian-code-satchin-panda-clock-health-science

Salgari: il ritorno a fumetti è di lusso - il mio articolo per Fumettologica

http://www.fumettologica.it/2018/07/sandokan-fumetti-star-comics/

Effetto nostalgia: Apple non cambierà il primo Store inaugurato da Steve Jobs - il mio articolo per La Stampa

http://www.lastampa.it/2018/07/23/tecnologia/effetto-nostalgia-apple-non-cambier-il-primo-store-inaugurato-da-steve-jobs-BzoAY5Qb6JwPHcmfltTbjP/pagina.html

Il sonno è una strana bestia. Perché ce n’è sempre troppo poco, soprattutto nel nostro mondo super stimolato e illuminato artificialmente. E la nostra percezione, in una economia del tempo libero (cioè fuori dall’orario di lavoro, quando diventiamo tutti consumatori e quindi è meglio se siamo svegli e spendiamo o guardiamo ma pubblicità) è enormemente falsata.

Money quote: “One 2014 study of more than 3,000 people in Finland found that the amount of sleep that correlated with the fewest sick days was 7.63 hours a night for women and 7.76 hours for men. So either that is the amount of sleep that keeps people well, or that’s the amount that makes them least likely to lie about being sick when they want to skip work. Or maybe people who were already sick with some chronic condition were sleeping more than that—or less—as a result of their illness. Statistics are tough to interpret. Isolated studies are tougher. “

https://www.theatlantic.com/magazine/archive/2017/01/how-to-sleep/508781/

Questo articolo è tornato a galla dopo un annetto o poco più che lo avevo citato. Lo riprendo per una seconda volta con piacere, perché il tema è forte: non tutti i big data sono tali. In particolare, qui si argomenta - con un interessante esperimento - l'idea che in molti casi si possa fare più e meglio con la riga di comando (perché la pipe parallelizza i calcoli su tutti i core) su un semplice laptop che non con i giganti del cloud con le soluzioni pensate per i big data. È anche una bella lezione su come ottimizzare una sequenza di comandi.

Money quote: "One especially under-used approach for data processing is using standard shell tools and commands. The benefits of this approach can be massive, since creating a data pipeline out of shell commands means that all the processing steps can be done in parallel. This is basically like having your own Storm cluster on your local machine. Even the concepts of Spouts, Bolts, and Sinks transfer to shell pipes and the commands between them. You can pretty easily construct a stream processing pipeline with basic commands that will have extremely good performance compared to many modern Big Data ™ tools."

https://adamdrake.com/command-line-tools-can-be-235x-faster-than-your-hadoop-cluster.html