October 2023 - nexa - server-nexa.polito.it

Re: [nexa] The Westminster Declaration (on the "Censorship Industrial Complex"). [CDT:L3]

by 380°

380° <g380(a)biscuolo.net> writes: > --8<---------------cut here---------------start------------->8--- [...] > Censorship robs us of the richness of life itself. Free speech is the > foundation for creating a life of meaning and a thriving humanity – > through art, poetry, drama, story, philosophy, song, and more. Song... si da il caso (caso?!?) che quasi esattamente 39 anni fa, il 12 Novembre 1984, fu pubblicato il quarto album degli Eurythmics dal titolo «1984 (For the Love of Big Brother)» [1], colonna sonora del film «Nineteen Eighty-Four» [2] dello stesso anno, tratto dal romanzo di George Orwell del... 1949 Il brano forse di maggior successo dell'album di cui sopra è «Sexcrime (Nineteen Eighty-Four)» [3], che a mo' di /esorcismo/ vi invito ad ascoltare _a_palla_ (con delle buone cuffie): https://yewtu.be/watch?v=DLevqpLSr14 Tra le poche parole, il testo recita: --8<---------------cut here---------------start------------->8--- I'll pull the bricks down One by one (Yeah!) Leave a big hole in the wall Just where you are looking in --8<---------------cut here---------------end--------------->8--- Nemmeno troppo velato riferimento a un'altra icona della musica rock: «The wall» dei Pink Floyd, del 1979 (30 anni dopo il romanzo 1984). Arte, poesia, storia, filosofia... canzoni. «Sono solo canzonette» (?!?) ...sono solo romanzi? Saluti, 380° [...] [1] https://en.wikipedia.org/wiki/1984_(For_the_Love_of_Big_Brother) [2] https://en.wikipedia.org/wiki/Nineteen_Eighty-Four_(1984_film) [3] https://en.wikipedia.org/wiki/Sexcrime_(Nineteen_Eighty-Four)_(song) -- 380° (Giovanni Biscuolo public alter ego) «Noi, incompetenti come siamo, non abbiamo alcun titolo per suggerire alcunché» Disinformation flourishes because many people care deeply about injustice but very few check the facts. Ask me about <https://stallmansupport.org>.

Oct. 18, 2023

The Westminster Declaration (on the "Censorship Industrial Complex"). [CDT:L7]

by 380°

Buongiorno, ...perché purtroppo sì: tocca /ancora/ ribadire l'ovvio. «“free” as in “free speech,” not as in “free beer.” » (https://www.gnu.org/philosophy/free-sw.en.html) https://www.severint.net/2023/10/18/dokumentiert-the-westminster-declaratio… --8<---------------cut here---------------start------------->8--- We write as journalists, artists, authors, activists, technologists, and academics to warn of increasing international censorship that threatens to erode centuries-old democratic norms. Coming from the left, right, and centre, we are united by our commitment to universal human rights and freedom of speech, and we are all deeply concerned about attempts to label protected speech as ‘misinformation,' ‘disinformation,' and other ill-defined terms. This abuse of these terms has resulted in the censorship of ordinary people, journalists, and dissidents in countries all over the world. Such interference with the right to free speech suppresses valid discussion about matters of urgent public interest, and undermines the foundational principles of representative democracy. Across the globe, government actors, social media companies, universities, and NGOs are increasingly working to monitor citizens and rob them of their voices. These large-scale coordinated efforts are sometimes referred to as the ‘Censorship-Industrial Complex.' This complex often operates through direct government policies. [Authorities in India] and [Turkey] have seized the power to remove political content from social media. The [legislature in Germany] and the [Supreme Court in Brazil] are criminalising political speech. In other countries, measures such as [Ireland's ‘Hate Speech' Bill], [Scotland's Hate Crime Act], the [UK's Online Safety Bill], and [Australia's ‘Misinformation' Bill] threaten to severely restrict expression and create a chilling effect. But the Censorship Industrial Complex operates through more subtle methods. These include visibility filtering, labelling, and manipulation of search engine results. Through deplatforming and flagging, social media censors have already silenced lawful opinions on topics of national and geopolitical importance. They have done so with the full support of ‘disinformation experts' and ‘fact-checkers' in the mainstream media, who have abandoned the journalistic values of debate and intellectual inquiry. As the Twitter Files revealed, tech companies often perform censorial ‘content moderation' in coordination with government agencies and civil society. Soon, the European Union's Digital Services Act will formalise this relationship by giving platform data to ‘vetted researchers' from NGOs and academia, relegating our speech rights to the discretion of these unelected and unaccountable entities. Some [politicians and NGOs] are even aiming to target end-to-end encrypted messaging apps like [WhatsApp, Signal, and Telegram]. If end-to-end encryption is broken, we will have no remaining avenues for authentic private conversations in the digital sphere. Although foreign disinformation between states is a real issue, agencies designed to combat these threats, such as the Cybersecurity and Infrastructure Security Agency in the United States, are increasingly being turned inward against the public. Under the guise of preventing harm and protecting truth, speech is being treated as a permitted activity rather than an inalienable right. We recognize that words can sometimes cause offence, but we reject the idea that hurt feelings and discomfort, even if acute, are grounds for censorship. Open discourse is the central pillar of a free society, and is essential for holding governments accountable, empowering vulnerable groups, and reducing the risk of tyranny. Speech protections are not just for views we agree with; we must strenuously protect speech for the views that we most strongly oppose. Only in the public square can these views be heard and properly challenged. What's more, time and time again, unpopular opinions and ideas have eventually become conventional wisdom. By labelling certain political or scientific positions as ‚misinformation‘ or ‚malinformation,‘ our societies risk getting stuck in false paradigms that will rob humanity of hard-earned knowledge and obliterate the possibility of gaining new knowledge. Free speech is our best defence against disinformation. The attack on speech is not just about distorted rules and regulations – it is a crisis of humanity itself. Every equality and justice campaign in history has relied on an open forum to voice dissent. In countless examples, including the abolition of slavery and the civil rights movement, social progress has depended on freedom of expression. We do not want our children to grow up in a world where they live in fear of speaking their minds. We want them to grow up in a world where their ideas can be expressed, explored and debated openly – a world that the founders of our democracies envisioned when they enshrined free speech into our laws and constitutions. The US First Amendment is a strong example of how the right to freedom of speech, of the press, and of conscience can be firmly protected under the law. One need not agree with the U.S. on every issue to acknowledge that this is a vital ‚first liberty‘ from which all other liberties follow. It is only through free speech that we can denounce violations of our rights and fight for new freedoms. There also exists a clear and robust international protection for free speech. The Universal Declaration of Human Rights (UDHR) was drafted in 1948 in response to atrocities committed during World War II. Article 19 of the UDHR states, ‚Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.‘ While there may be a need for governments to regulate some aspects of social media, such as age limits, these regulations should never infringe on the human right to freedom of expression. As is made clear by Article 19, the corollary of the right to free speech is the right to information. In a democracy, no one has a monopoly over what is considered to be true. Rather, truth must be discovered through dialogue and debate – and we cannot discover truth without allowing for the possibility of error. Censorship in the name of ‚preserving democracy‘ inverts what should be a bottom-up system of representation into a top-down system of ideological control. This censorship is ultimately counter-productive: it sows mistrust, encourages radicalization, and de-legitimizes the democratic process. In the course of human history, attacks on free speech have been a precursor to attacks on all other liberties. Regimes that eroded free speech have always inevitably weakened and damaged other core democratic structures. In the same fashion, the elites that push for censorship today are also undermining democracy. What has changed though, is the broad scale and technological tools through which censorship can be enacted. We believe that free speech is essential for ensuring our safety from state abuses of power – abuses that have historically posed a far greater threat than the words of lone individuals or even organised groups. For the sake of human welfare and flourishing, we make the following 3 calls to action. • We call on governments and international organisations to fulfill their responsibilities to the people and to uphold Article 19 of the UDHR. • We call on tech corporations to undertake to protect the digital public square as defined in Article 19 of the UDHR and refrain from politically motivated censorship, the censorship of dissenting voices, and censorship of political opinion. • And finally, we call on the general public to join us in the fight to preserve the people's democratic rights. Legislative changes are not enough. We must also build an atmosphere of free speech from the ground up by rejecting the climate of intolerance that encourages self-censorship and that creates unnecessary personal strife for many. Instead of fear and dogmatism, we must embrace inquiry and debate. We stand for your right to ask questions. Heated arguments, even those that may cause distress, are far better than no arguments at all. Censorship robs us of the richness of life itself. Free speech is the foundation for creating a life of meaning and a thriving humanity – through art, poetry, drama, story, philosophy, song, and more. This declaration was the result of an initial meeting of free speech champions from around the world who met in Westminster, London, at the end of June 2023. As signatories of this statement, we have fundamental political and ideological disagreements. However, it is only by coming together that we will defeat the encroaching forces of censorship so that we can maintain our ability to openly debate and challenge one another. It is in the spirit of difference and debate that we sign the Westminster Declaration. [List of Signatories] References: [Authorities in India] <http://slate.com/technology/2023/04/twitter-blocked-pakistan-india-modi-mus…> [Turkey] <http://www.washingtonpost.com/technology/2023/05/13/turkey-twitter-musk-erd…> [legislature in Germany] <https://www.dw.com/en/germany-criminalizes-denying-war-crimes-genocide/a-63…> [Supreme Court in Brazil] <http://apnews.com/article/jair-bolsonaro-brazil-government-af5987e833a681e6…> [Ireland's ‘Hate Speech' Bill] <http://www.telegraph.co.uk/world-news/2023/06/17/irish-people-jailed-hate-s…> [Scotland's Hate Crime Act] <https://www.economist.com/the-world-ahead/2021/11/08/scotlands-new-hate-cri…> [UK's Online Safety Bill] <http://techcrunch.com/2023/07/05/uk-online-safety-bill-risks-e2ee> [Australia's ‘Misinformation' Bill] <http://www.abc.net.au/news/2023-06-25/fines-to-punish-online-misinformation…> [politicians and NGOs] <http://meedan.com/project/cryptochat> [WhatsApp, Signal, and Telegram] <http://echcrunch.com/2023/07/05/uk-online-safety-bill-risks-e2ee/> [List of Signatories] <https://www.severint.net/2023/10/18/dokumentiert-unterzeichner-der-westmins…> --8<---------------cut here---------------end--------------->8--- Cordiali saluti, 380° -- 380° (Giovanni Biscuolo public alter ego) «Noi, incompetenti come siamo, non abbiamo alcun titolo per suggerire alcunché» Disinformation flourishes because many people care deeply about injustice but very few check the facts. Ask me about <https://stallmansupport.org>.

Oct. 18, 2023

Re: [nexa] IA, lavoro, immaginari [CDT:L3] [CDT:L5] [CDT:L-1]

by Maria Chiara Pievatolo

On 14/10/23 10:23, 380° wrote: > > Non so come sia possibile ma a me questa analisi critica kantiana è > /intuitivamente/ chiara, direi scontata: un po' come dovrebbe essere > scontato a qualsiasi geografo/astrofisico che la mappa del pianeta terra > NON è il pianeta terra [1], analogamente a un informatico dovrebbe > essere chiaro che il modello usato nell'algoritmo implementato nel > programma [2] non corrisponde all'"oggetto" studiato, a meno che non ci > si /limiti/ a fare lo spelling di "sugar" o a calcolare "2 + 2" (numeri > in base 10) [1]. > > In altre perole, non *deve* bastare il "giochino" di dire che un certo > modello dell'universo è giusto e un'altro sbagliato con giustificazini > "sensibili", occorre saper spiegare _perché_ e, di conseguenza, fare in > modo che _altri_ possano liberamente /giudicare/ qual'è quello > relativamente più sbagliato. [1] > Questa è anche la differenza fra scienza e opinione corretta che si ritrova anche in Platone https://btfp.sp.unipi.it/dida/menone/ar01s19.xhtml - tanto è vero che, per alcuni aspetti, Kant potrebbe anche essere trattato come un platonico moderno (*) Che la realtà sia diversa dai "dati" e dalle elaborazioni che ne facciamo non è per tutti così scontato. Come ha ricordato Guido Vetere, nel 2008 (https://www.wired.com/2008/06/pb-theory/) qualcuno ha proclamato che quando abbiamo molti dati e una potenza di calcolo abbastanza grande andare alla ricerca delle cause non serve più: bastano le correlazioni. Per capire quanto sia tolemaico questo modo di pensare non occorre nemmeno scomodare Kant, ma è sufficiente David Hume (a cui Kant riconosceva il merito di averlo svegliato dal suo sonno dogmatico): dalla circostanza che siamo abituati a vedere B seguire ad A perché ne abbiamo fatto esperienza tantissime volte non possiamo derivare che A è causa di B. L'esperienza ci dice soltanto che *finora* abbiamo visto A e B in correlazione. Non ci può dire, come esperienza, che ciò avverrà anche un domani che non abbiamo ancora esperito.[**) Le nostre abitudini sono solo le nostre abitudini, non la struttura della realtà: pare ovvio, ma moltissimi sono andati dietro a Chris Anderson. E analogamente, il fatto che io sia in grado di *imitare* con la una tecnica certi fenomeni e comportamenti non significa che li abbia compresi.(***) I fratelli Lumière, con la loro tecnica, *imitarono* in modo così convincente l'arrivo di un treno da far scappare gli spettatori dalla sala. Avevano capito il funzionamento della locomotiva a vapore? No, avevano inventato il cinematografo. L'informatico assolutista si rende conto che saper *imitare* qualcosa in modo da indurre i più a *interpretare* la sua imitazione come se fosse la cosa stessa non significa, di per sé, averla anche compresa? Weizenbaum, già con ELIZA, notò che la settorializzazione della nostra cultura e del nostro modo di trasmetterla era tale che la risposta più probabile a questa domanda era "no". Buonanotte, MCP (*)Per farsi un'idea della differenza fra Kant e Platone bisogna guardare due passi della Critica della ragion pura, uno nell'Introduzione (parola chiave: "colomba") e l'altro nel primo libro della Dialettica trascendentale (Sezione I. Delle idee in generale), ma andrei molto fuori tema. (**) N. Taleb, riproponendo qualcosa di filosoficamente ben noto, ha scritto un libro di successo proprio sul cigno nero che da solo distrugge la correlazione abituale fra i cigni e il colore bianco. (***) L'imitazione (mimesis) tramite la tecnica produce sapere? Secondo Platone, no: https://btfp.sp.unipi.it/dida/resp/ar01s48.xhtml

Oct. 18, 2023

Re: [nexa] "Why only 1% of the Snowden Archive will ever be published" [CDT:L-1]

by 380°

Buongiorno, andate a chiedere alla intelligence dei *cattivi* se non hanno almeno il 50% dei file non pubblicati :-O Quello non è materiale per giornalisti, è materiale per /archeologi/. “The main reason for only a small percentage [...] was diminishing interest.” mavaff... "J.C. DE MARTIN" <juancarlos.demartin(a)polito.it> writes: > *Why only 1% of the Snowden Archive will ever be published > * > /Speaking to Computer Weekly after we published new revelations from the > Snowden archive, the Guardian’s Pulitzer Prize winner, Ewen MacAskill, > explains why more of the Snowden trove is unlikely to see the light of > day/ > > By Stefania Maurizi, Published: 11 Oct 2023 10:30 [...] --8<---------------cut here---------------start------------->8--- “The bottom line is that Snowden is facing charges under the Espionage Act. If he was ever to return to the US and face trial, the documents could be used against him. All journalists have a duty to protect source material. How best to do that? How long would The New York Times be willing to store them? Where else could they be stored? Should the documents be destroyed?” --8<---------------cut here---------------end--------------->8--- Ecco appunto, è inutile che ci girano (giriamo?) in giro, è ovvio il motivo per il quale pochissimo del materiale è stato pubblicato, si chiama "Effetto Julian Assange & Co." ...non appena la soglia di pericolo di essere sottoposti a simili trattamenti scenderà sollo il livello "normale", ne vedremo delle belle. > continua qui: > https://www.computerweekly.com/news/366554957/Why-only-1-of-the-Snowden-Arc… É la seconda parte di questo articolo: «New revelations from the Snowden archive surface» /A decade after Snowden exposed NSA’s mass surveillance in cooperation with the British GCHQ, only about 1% of the documents have been published – but three major facts can finally be revealed thanks to a doctoral thesis in applied cryptography by Jacob Appelbaum/ By Stefania Maurizi, Published: 19 Sep 2023 21:58 https://www.computerweekly.com/news/366552520/New-revelations-from-the-Snow… Toh: Jacob Applebaum, il molestatore sessuale seriale [1]!... deve essere una malattia trasmissibile dei "top manager" di Wikileaks e dintorni... oppure "Gomblotto" come piace dire al FAT CHECKER quadratico medio!?! Poi dopo un periodo *quasi* sottotraccia, salta fuori bello bello con la sua tesi di dottorato «Communication in a world of pervasive surveillance - Sources and methods: Counter-strategies against pervasive surveillance architecture» [2] con avisors Tanja Lange, D.J. Bernstein [3] L'articolo racconta che nella tesi Applebaum fa tre nuove rilelazioni, una delle quali non è per nulla nuova quindi chissenefrega: --8<---------------cut here---------------start------------->8--- 1. The NSA listed Cavium, an American semiconductor company marketing Central Processing Units (CPUs) – the main processor in a computer which runs the operating system and applications – as a successful example of a “SIGINT-enabled” CPU supplier. Cavium, now owned by Marvell, said it does not implement back doors for any government. 2. The NSA compromised lawful Russian interception infrastructure, SORM. The NSA archive contains slides showing two Russian officers wearing jackets with a slogan written in Cyrillic: “You talk, we listen.” The NSA and/or GCHQ has also compromised Key European LI [lawful interception] systems. [...] “The NSA’s successful cryptographic enabling is by definition the introduction of intentional security vulnerabilities that they are then able to exploit, and they do exploit them often in an automated fashion to spy,” he said. “One such method is sabotaging a secure random generator.” [...] Will the 99% of the Snowden archive ever be published? How many important revelations like these do the unpublished documents still contain? It is impossible to say so long as the archive remains unpublished. It is also unclear how many copies of the full archive remain available and who has access to them. [...] it was a terrible day when The Guardian allowed GCHQ to destroy the copy of the archive in the UK. [...] “A copy of the Snowden documents remains locked in an office at The Times, as far as I know.” [...] We asked The Intercept whether the publication is still in possession of the Snowden file. A spokesperson replied: “The Intercept does not discuss confidential news-gathering materials.” --8<---------------cut here---------------end--------------->8--- La rivelazione 2. non è altro che la conferma che ECHELON è vivo e vegeto e ha solo cambiato nome. La 1. è la conferma che le backdoors della _crittografia_ *sono* implementate anche nei microprocessori, e questo è probabilmente uno dei motivi per il quale il 99% degli Snowden files sono gelosamente mantenuti segreti... come i segreti di pulcinella. Security by obscurity è una boiata, lo sanno tutti... ma sono talmente OLTRE la disperazione che adesso non possono fare altro che /disperatamente/ tentare di mantenere segrete le backdoors nella speranza che non siano sfruttate dai *cattivi*. Lo dico in un'altro modo, così forse si capisce meglio: secondo voi quanto ci metteranno i *cattivi* a compromettere i sistemi chiave dei *buoni*? Saluti, 380° [1] https://en.wikipedia.org/wiki/Jacob_Appelbaum#Allegations_of_sexual_miscond… [2] https://pure.tue.nl/ws/portalfiles/portal/197416841/20220325_Appelbaum_hf.p… 355 pagine in PDF A4 [3] sì, quello di "Bernstein v. United States" del 1995 sulla costituzionalità delle norme USA in merito alla crittografia, caso permanentemente "on hold" https://cr.yp.to/export/2003/10.15-bernstein.txt ...perché gli USA sono uno stato di diritto "sulla parola". P.S.: e dire che per un breve periodo durante i miei studi in materia ho temuto, anche grazie all'insopportabile pressione ambientale, di essere affetto da complottismo :-D -- 380° (Giovanni Biscuolo public alter ego) «Noi, incompetenti come siamo, non abbiamo alcun titolo per suggerire alcunché» Disinformation flourishes because many people care deeply about injustice but very few check the facts. Ask me about <https://stallmansupport.org>.

Oct. 17, 2023

Re: [nexa] Doctorow, "Why Big Tech, Cops, and Spies Were Made for One Another"

by Angelo Raffaele Meo

In Italia il GPDP, che ha inflitto una multa pesantissima alla Bocconi per aver svolto esami remoti con un prodotto poco sicuro. ha ignorato il fatto che la maggioranza delle scuole italiane per anni ha continuato a usare ben noti prodotti vietati dalle norme di legge comunitarie e nazionali. ________________________________ From: nexa <nexa-bounces(a)server-nexa.polito.it> on behalf of J.C. DE MARTIN <juancarlos.demartin(a)polito.it> Sent: Monday, October 16, 2023 2:27 PM To: Nexa <nexa(a)server-nexa.polito.it> Subject: [nexa] Doctorow, "Why Big Tech, Cops, and Spies Were Made for One Another" Why Big Tech, Cops, and Spies Were Made for One Another The American surveillance state is a public-private partnership. Cory Doctorow (16 dicembre 2023) (Cory Doctorow’s latest book is “The Internet Con: How to Seize the Means of Computation.”) The techlash has finally reached the courts. Amazon’s in court. Google’s in court. Apple’s under EU investigation. The French authorities just kicked down Nvidia’s doors and went through their files looking for evidence of crimes against competition. People are pissed at tech: about moderation, about monopolization, about price gouging, about labor abuses, and — everywhere and always — about privacy. >From experience, I can tell you that Silicon Valley techies are pretty sanguine about commercial surveillance: “Why should I care if Google wants to show me better ads?” But they are much less cool about government spying: “The NSA? Those are the losers who weren’t smart enough to get an interview at Google.” And likewise from experience, I can tell you that government employees and contractors are pretty cool with state surveillance: “Why would I worry about the NSA spying on me? I already gave the Office of Personnel Management a comprehensive dossier of all possible kompromat in my past when I got my security clearance.” But they are far less cool with commercial surveillance: “Google? Those creeps would sell their mothers for a nickel. To the Chinese.” What are they both missing? That American surveillance is a public-private partnership: a symbiosis between a concentrated tech sector that has the means, motive, and opportunity to spy on every person in the world and a state that loves surveillance as much as it hates checks and balances. Big Tech, cops, and surveillance agencies were made for one another. [...] continua qui: https://theintercept.com/2023/10/16/surveillance-state-big-tech/

Oct. 16, 2023

Re: [nexa] Doctorow, "Why Big Tech, Cops, and Spies Were Made for One Another"

by Roberto Resoli

Il 16 ottobre 2023 14:27:14 CEST, "J.C. DE MARTIN" <juancarlos.demartin(a)polito.it> ha scritto: >*Why Big Tech, Cops, and Spies Were Made for One Another >* >/The American surveillance state is a public-private partnership. >/ Lo diceva Bruce Schneier nel 2013: "Surveillance is the business model of the internet, and business surveillance gives goverments access to data they couldn't get otherwise" https://i.redsnake.io/watch?v=h0d_QDgl3gI&t=190s rob

Oct. 16, 2023

Doctorow, "Why Big Tech, Cops, and Spies Were Made for One Another"

by J.C. DE MARTIN

*Why Big Tech, Cops, and Spies Were Made for One Another * /The American surveillance state is a public-private partnership. / Cory Doctorow (16 dicembre 2023) /(Cory Doctorow’s latest book is “The Internet Con: How to Seize the Means of Computation.”)/ The techlash has finally reached the courts. Amazon’s in court. Google’s in court. Apple’s under EU investigation. The French authorities just kicked down Nvidia’s doors and went through their files looking for evidence of crimes against competition. People are pissed at tech: about moderation, about monopolization, about price gouging, about labor abuses, and — everywhere and always — about privacy. From experience, I can tell you that Silicon Valley techies are pretty sanguine about commercial surveillance: “Why should I care if Google wants to show me better ads?” But they are much less cool about government spying: “The NSA? Those are the losers who weren’t smart enough to get an interview at Google.” And likewise from experience, I can tell you that government employees and contractors are pretty cool with state surveillance: “Why would I worry about the NSA spying on me? I already gave the Office of Personnel Management a comprehensive dossier of all possible kompromat in my past when I got my security clearance.” But they are far less cool with commercial surveillance: “Google? Those creeps would sell their mothers for a nickel. To the Chinese.” What are they both missing? That American surveillance is a public-private partnership: a symbiosis between a concentrated tech sector that has the means, motive, and opportunity to spy on every person in the world and a state that loves surveillance as much as it hates checks and balances. Big Tech, cops, and surveillance agencies were made for one another. [...] continua qui: https://theintercept.com/2023/10/16/surveillance-state-big-tech/

Oct. 16, 2023

Re: [nexa] 4M€: pochi? Tanti? dettagli, please....

by Damiano Verzulli

Il 16/10/23 12:28, Giuseppe Attardi ha scritto: > L’ho detto perché sono pochi: l’obiettivo del progetto è di produrre > sostanzialmente due cose: > > 1. Una raccolta di testi per l’apprendimento di LLM > 2. Centinaia (100s) di modelli allenati (anche se il GANTT ne indica > solo uno da completare in 18 mesi, anche se il dataset sarà > disponibile dopo 12!) aiuterebbe avere il testo "ufficiale" del progetto... dato che la differenza fra "100s" e "1" non è trascurabile... Sarebbe utile anche capire perché... se l'obiettivo è "palesemente impossibile" (come mi pare di intuire da quello che lei scrive), qualcuno, in EU, si sia prodigato per "buttare" 3.880.687,50 € [1] in questo progetto. In ogni caso, intuisco che siamo di fronte a un muro invalicabile: da parte mia, il sostenere che con 4M€ si possono fare UN SACCO di cose, come appunto il progetto HPLT dice di fare. Dall'altro, chi preferisce dire: "Servono piu' soldi. Per fare ChatGPT-4, 5, 6, 894... servono piu' soldi. Dateci i soldi" Temo che fin quando non si riuscira' a convincere qualcuno come me di quest'ultima tesi... sara' certamente impossibile convincere chi, quei soldi, dovrebbe cacciarli realmente (aka: la politica). Mi viene in mente Quintarelli, che citava Andreotti: "Non basta avere ragione. Bisogna che te la diano..." Mi fermo qua. Ho gia' abusato troppo del tempo di tutti. Saluti, DV [1] https://cordis.europa.eu/project/id/101070350 -- Damiano Verzulli e-mail:damiano@verzulli.it --- possible?ok:while(!possible){open_mindedness++} --- "...I realized that free software would not generate the kind of income that was needed. Maybe in USA or Europe, you may be able to get a well paying job as a free software developer, but not here [in Africa]..." -- Guido Sohne - 1973-2008 http://ole.kenic.or.ke/pipermail/skunkworks/2008-April/005989.html

Oct. 16, 2023

Re: [nexa] IA, lavoro, immaginari

by Guido Vetere

comincio a rivalutare gli embedding :-))) On Mon, 16 Oct 2023 at 12:09, Giacomo Tesio <giacomo(a)tesio.it> wrote: > Il giorno Mon, 16 Oct 2023 11:48:19 +0200 Guido Vetere ha scritto: > > > noi ora qui possiamo metterci d'accordo su cosa intendiamo per > > 'intelligenza', e cosa in particolare per 'intelligenza aliena' > > Certo che potremmo! :-) > > E in fondo, se definissimo come "rotazione aliena" quella che il sole ed > il resto dell'universo compiono intorno alla terra, potremmo ben dire > che il sole e l'universo tutto ruotano intorno alla terra! > > Questa "rotazione aliena" continuerebbe testardamente a non esistere... > ma che importa? > > > E perché non ridefinire "libertà"? > Perché non ridefinire "conoscenza"? > Perché non ridefinire "verità"? > > In fondo, i transumanisti stanno cercando letteralmente di ridefinire > "vita", sperando di spingere un numero sufficiente di persone al > suicidio assistito attraverso allucinazioni come queste... > > > Giacomo >

Oct. 16, 2023

Re: [nexa] 4M€: pochi? Tanti? dettagli, please....

by Giuseppe Attardi

L’ho detto perché sono pochi: l’obiettivo del progetto è di produrre sostanzialmente due cose: 1. Una raccolta di testi per l’apprendimento di LLM 2. Centinaia (100s) di modelli allenati (anche se il GANTT ne indica solo uno da completare in 18 mesi, anche se il dataset sarà disponibile dopo 12!) Il costo di allenare un LLM come GPT è stimato intorno ai $100 milioni. Fai il conto tu di quanti se ne possono allenare con 4 milioni (costo totale, non contributo UE). 4 milioni su tre anni per 8 partner fa 166.000€ a testa, sufficienti a pagare circa 3 ricercatori a partner. La lista dei contributori a GPT-4 ha oltre 300 firme: https://openai.com/contributions/gpt-4 La scala è quella dei progetti del CERN. — Beppe > On 16 Oct 2023, at 09:02, <nexa-request(a)server-nexa.polito.it> <nexa-request(a)server-nexa.polito.it> wrote: > > From: Damiano Verzulli <damiano(a)verzulli.it <mailto:damiano@verzulli.it>> > To: nexa(a)server-nexa.polito.it <mailto:nexa@server-nexa.polito.it> > Subject: [nexa] 4M€: pochi? Tanti? dettagli, please.... [Was: <IA, > lavoro, immaginar>] > Message-ID: <d98273cd-ccf1-4943-9831-bb162d15b30a(a)verzulli.it <mailto:d98273cd-ccf1-4943-9831-bb162d15b30a@verzulli.it>> > Content-Type: text/plain; charset="utf-8"; Format="flowed" > > Il 16/10/23 04:59, Giuseppe Attardi ha scritto: >> [...] >> Peccato che il finanziamento totale sia di 4 milioni €, con cui ci si >> fa poco. >> [...] > > 4 milioni di €, *PUBBLICI*, sono una quantita' *ENORME* di denaro. > > Mi sanguina il cuore ad osservare la "leggerezza" con la quale tale > somma di denaro viene sminuita. > > Perché sono pochi? Cosa cambierebbe, rispetto al progetto HPLT, se > fossero di piu'? Ma, soprattutto, cosa "torna indietro", precisamente, > HPLT, a fronte di quei 4M€.... soprattutto in considerazione del fatto > che mi aspetto che i 4M€ sia una percentuale alta di quello che loro > stessi hanno chiesto? > Detto in altri termini: non solo i 4M€ a HPLT sono tantissimi... ma io, > terzo, voglio pure capire BENISSIMO cosa ci hanno fatto / stanno > facendo... (il sito [1], fortunatamente, aiuta)

Oct. 16, 2023