February 2023 - nexa - server-nexa.polito.it

Il lavoro sporco di ChatGPT (intervista in spagnolo)
by Antonio Casilli Feb. 27, 2023

Feb. 27, 2023

Segnalo questa intervista che ho rilasciato al quotidiano El Mercurio. Si parla di ChatGPT e insegnamento, arte (in riferimento al pensiero del filosofo spagnolo Daniel Innerarity), lavoro e di come solo sociopatici arricchiti come Bezos e Musk vorrebbero veramente dell'AGI. Entrevista - Lo humano y lo artificial Conversar con el ChatGPT: ¿qué hay detrás de la pantalla? Juan Rodríguez M., El Mercurio, 26 febrero 2023 https://digital.elmercurio.com/2023/02/26/E/4647PB39 (paywalled) Il testo in spagnolo (per facilitar lettura e traduzione), è disponibile qui: https://www.casilli.fr/2023/02/26/chatgpt-y-el-trabajo-sucio-de-lia-entrevi… Cheers, ---a -- Antonio A. Casilli Professor, Telecom Paris | Institut Polytechnique de Paris Member, Interdisciplinary Institute for Innovation (CNRS) Associate Member, LACI-LAP (EHESS) Associate researcher, Weizenbaum-Institut, Berlin Member, Scholarly council UCLA Center for Critical Internet Inquiry (C2i2) Faculty Fellow, Nexa Center for Internet & Society *We respect your right to disconnect. My workflow efficiency accounts for this email send time. Taking action or responding outside of office hours is not mandatory*

1 0

Sempre su Chat GPT
by Diego Giorio Feb. 26, 2023

Feb. 26, 2023

Condivido una ricerca fatta oggi. Pure si scusa per essere stato beccato in fallo. Io continuo a stupirmi di come mescoli risposte accurate ed una capacità di dialogo apparentemente umana con scivoloni colossali. I primi software di traduzione degli anni '90 potevano avere frasi più o meno riuscite, ma l'approssimazione era abbastanza costante. Questa va da una risposta precisa al considerare reale l'unicorno. Buona settimana a tutti. D. quando fu coniata la prima medaglia di san benedetto? La prima medaglia di San Benedetto fu coniata nel 1880. La medaglia fu creata da padre Benedetto della Croce, un monaco benedettino italiano, che voleva diffondere la devozione a San Benedetto, il fondatore dell'ordine benedettino. La medaglia è diventata molto popolare tra i fedeli cattolici e viene spesso indossata come simbolo di protezione e benedizione. [X][dgiorio(a)hotmail.com] No, questa risposta si riferisce alla versione attuale, che però si basa su medaglie più antche Chiedo scusa per l'errore nella mia precedente risposta. Infatti, la medaglia di San Benedetto si basa su medaglie più antiche che risalgono almeno al XVII secolo. La prima medaglia con l'immagine di San Benedetto fu coniata a Norcia, l'Italia, nel 1647. Questa medaglia era molto simile alla versione attuale, ma invece di mostrare l'immagine di San Benedetto sul fronte, mostrava un monogramma che rappresentava le iniziali del motto latino di San Benedetto: "Crux Sancti Patris Benedicti" (La Croce del Santo Padre Benedetto). La medaglia di San Benedetto come la conosciamo oggi è stata sviluppata nel corso del tempo, con aggiunte e modifiche apportate all'immagine e al testo presenti sulle medaglie. Tuttavia, il significato e la protezione associati alla medaglia rimangono gli stessi di quelli delle prime medaglie coniate in onore di San Benedetto. Regenerate response

1 0

Google adverts direct pregnant women to services run by UK anti-abortion groups | Abortion | The Guardian
by Alberto Cammozzo Feb. 25, 2023

Feb. 25, 2023

<https://www.theguardian.com/world/2023/feb/25/google-adverts-direct-pregnan…> Women seeking online advice about abortions are being directed to pregnancy counselling services run by anti-abortion campaigners, an Observer investigation has found. Google adverts that are styled to look like real search results and appear above genuine listings are routinely being shown to people searching key terms relating to pregnancy and abortion. In an analysis this month, 117 out of 251 adverts shown by Google UK to a user searching 40 key phrases, including “NHS abortion advice”, “confidential abortion support” and “pregnant teenager help”, were from groups opposed to abortion. The findings reveal the marketing efforts of anti-abortion groups in the UK and have led to concerns that women could be exposed to biased information when seeking out medical advice. One sexual health charity described the advertising as “clearly immoral”. The adverts in the analysis – which were offered to a woman in her 20s in London in early February – contain a small tag marking them as advertising but look similar to real search results and appear above trusted information sources, including the NHS website. In some cases they promote advice services that claim to offer impartial support but do not clearly state the anti-abortion views of the people behind them. One of the biggest advertisers was Pregnancy Crisis Helpline, whose adverts appeared after searches for 14 out of the 40 phrases tested in the analysis. They included one that said: “Considering an abortion? Talk to someone,” and appeared after searches for “buy abortion pill” and “pregnant teenager help”. People clicking on the links were taken to the helpline’s website, which says it is a “safe and confidential place” offering “support for women struggling with an unplanned pregnancy”. But while it says it does “not refer for abortions”, neither does it offer information about its organisers’ anti-abortion views and presents itself as an impartial service offering support “away from all the pressures”. In reality, the helpline was co-launched with Christian Concern, a rightwing evangelical organisation that wants abortion to be banned. The helpline’s trustees include Regan King, a pastor at the Angel church in Islington, London, who has described abortion as “Disgusting. Disturbing. Grim. Gruesome. Horrifying. Shocking. Terrible. Vile” and likened it to “the new slave trade”. Another trustee is Christian Clive Copus, a former director of the anti-abortion campaign group the Prolife Alliance. The helpline has recently reported a spike in the number of people contacting it, saying it had 2,000 clients in 2022 compared with 500 in 2021. Other Google adverts directed women to a counselling service run by the charity Life, which opposes abortion. One said: “We provide a safe space for you to explore your feelings away from outside pressures. Talk through your options in confidence.” Google said the adverts flagged by the Observer complied with its rules, highlighting that they contain an “ad” tag in bold lettering and a line stating that the services they promote do not provide abortions. But Lisa Hallgarten, head of policy at young people’s sexual health charity Brook, described the marketing approach as “clearly immoral” and said adverts provided to those searching terms such as “NHS abortion advice” could delay women’s access to healthcare. “We are really concerned that people looking for impartial support are being directed to organisations and websites where they could experience the complete opposite,” she said. Pam Lowe, an expert on anti-abortion activism, said: “Anyone who puts ‘NHS abortion’ into Google should be shown a link to the NHS website as the first result. There’s a risk that people could end up with biased information.” Labour MP Stella Creasy, who has campaigned on access to reproductive healthcare, urged the government to compel tech companies to remove potentially harmful listings. She said there was a difference between allowing free speech and “seeking to mislead vulnerable readers who need to be confident that what they are reading is medically sound and impartial”. The Pregnancy Crisis Helpline said it stated on its website that it did not offer medical advice or “refer for abortion”. “If a client asks for medical advice, we signpost them to their GP, 111 or A&E as appropriate,” said Toby Cosh, chair of the trustees. He added that while the helpline had been set up with support from Christian Concern, it had operated independently since 2018 and was a registered UK charity. Kerry Smart, chief executive at Life, said: “Our person-centred online pregnancy listening service was inspired by the Samaritans and is non-coercive and non-judgmental.” She added that people using Life’s “listening services” were told it did not refer for abortion or give information on abortion providers, and that the charity complied with ethical guidelines from the British Association for Counselling and Psychotherapy. “If clients wish to explore medical topics, we inform – not advise – using NHS information in skilled listening sessions,” she said. “We agree that attempts to mislead or give false information are wrong.” Other adverts that appear on UK Google searches relating to pregnancy and abortion came from regulated abortion providers, including MSI Reproductive Choices and the British Pregnancy Advisory Service. They appeared alongside adverts placed by anti-abortion charities in the UK and US that do not run counselling services and instead direct people to webpages about the ethics of abortion. The Observer understands that they do not always pay for their adverts because of a Google scheme that means organisations with charity status can be granted free advertising credits. Google said: “We know that people come to Google looking for information they can trust, and we’ve invested heavily in providing a safe and transparent experience. “When it comes to abortion-related ads, we require an added level of transparency so that people seeking abortion-related resources know what services an advertiser actually provides. Any organisation that wants to target queries related to getting an abortion must complete our certification process and clearly disclose whether they do or do not offer abortions.”

1 0

[CFP] The Data EConomy (DEC) Workshop 2023
by Martino Trevisan Feb. 24, 2023

Feb. 24, 2023

[Apologies if you receive multiple copies of this mail] The Data EConomy (DEC) Workshop 2023 in conjunction with SIGMOD 2023 Seattle, June 18, 2023 Website: https://sites.google.com/view/data-economy-2023/ ORGANIZERS AND CHAIRS * Nikolaos Laoutaris, IMDEA Networks Institute * Georgia Koutrika, Athena Research Center WORKSHOP DESCRIPTION Data-driven decision making powered by Machine Learning (ML) algorithms is changing how the society and the economy work and is having a profound positive impact on our daily life. With the exception of very large companies that have both the data and the skills to develop powerful ML-driven services, the large majority of provably possible ML services, from e-health, to transportation and predictive maintenance, to name just a few, still remain at the idea or prototype level for the simple reason that data, the skills to manipulate them, and the business models to bring them to market, seldom co-exist under the same roof. The value of data comes from its contextualisation and combination with other data. Indeed, this can give way to many new services and products. Furthermore, data has to somehow meet with the ML and business skills that can unleash its full power for the society and economy. This has given rise to a highly dynamic sector around the Data Economy, involving Data Providers/Controllers, data Intermediaries, often-times in the form of Data Marketplaces or Personal Information Management Systems for end-users to control and even monetise their personal data. Despite its huge potential and observed initial growth, the Data Economy is still at its nascent phase and, therefore, faces a yet uncertain future and a series of existential challenges. Such challenges include a broad range of technical matters across multiple disciplines of Computer Science including databases, machine learning, distributed systems, security and privacy, and human computer interaction. TOPICS The mission of the Data Economy workshop will be to bring together all the CS skills required for helping the Data Economy liftoff by addressing a range of technical challenges including, but are not limited, to the ones below: * Design, architecture, systems and protocols for Data Marketplaces, Data Vendors, and Personal Information Management Systems (PIMS) * Consent management and taxonomy of data processing purposes * Sending the data to the algorithm vs. sending the algorithm to the data * Federated and distributed learning in the data economy * Data management and querying in Data Marketplaces * Secure data exchange and delivery mechanisms * Federated data catalogues and data discovery mechanisms * Heterogeneous and federated DBMS, metadata management * Information Integration and Data Quality * Privacy/data protection and the data economy * Data pricing mechanisms for individual and aggregated data * How to buy data – data purchase policies and algorithms * Protecting data ownership rights for commercial datasets * NFTs, blockchains, smart contracts and their role in the data economy * Trusted execution, cloud computing, distributed storage and their role in the data economy * Data representation and exchange standards in the data economy * Understanding the value of data in different applications and domains and across the data value chain * Cryptographic approaches, including FHE, SMPC, DP, and their role and limits in the data economy * UX and HCI challenges for data marketplaces and PIMS * Understanding and decoding the Terms & Conditions of data marketplaces and data transactions * Federation and interoperability standards and protocols in the data economy * Measurement studies related to the data economy * Large-scale experiments and validation studies for the data economy SUBMISSION DETAILS We solicit papers up to 6 pages long, and demonstrations (up to 4 pages long). Submissions will include author names, hence allowing single-anonymous reviewing. Submission Website: https://acm-data-economy23.hotcrp.com/ IMPORTANT DATES * Paper Submission: March 10, 2023 * Notification of Acceptance: April 15, 2023 * Camera-ready Submission: April 30, 2023 * Workshop Date: June 18, 2023 (Sunday) * Workshop format: full-day

1 0

Ho fatto lezione con ChatGPT. So what?
by Antonio Casilli Feb. 23, 2023

Feb. 23, 2023

L'altroieri ho fatto lezione. A quanto pare è oramai cosa abbastanza rara perché la stampa francese ne parli... https://www.leparisien.fr/etudiant/examens/un-prof-lance-un-chatgpt-challen… -- Antonio A. Casilli Professor, Telecom Paris | Institut Polytechnique de Paris Member, Interdisciplinary Institute for Innovation (CNRS) Associate Member, LACI-LAP (EHESS) Associate researcher, Weizenbaum-Institut, Berlin Member, Scholarly council UCLA Center for Critical Internet Inquiry (C2i2) Faculty Fellow, Nexa Center for Internet & Society *We respect your right to disconnect. My workflow efficiency accounts for this email send time. Taking action or responding outside of office hours is not mandatory*

3 4

Lo SPID sta finendo, e un decennio se ne va.
by 380° Feb. 23, 2023

Feb. 23, 2023

Buonasera, per chi non lo conoscesse, avviso che l'autore, Uriel Fanelli, ha uno stile /turpilo-barocco/, ma con un adeguato filtro e tanta pazienza forse qualcosa del contenuto si riesce a capire ed eventualmente a criticare. Siccome so che in questa lista ci sono persone molto bene informate sui fatti, chiedo loro se vogliono commentare. https://keinpfusch.net/lo-spid-sta-finendo/ febbraio 23, 2023 «Lo SPID sta finendo, e un decennio se ne va.» --8<---------------cut here---------------start------------->8--- Parlare di digitalizzazione e' sempre un pericolo, per una ragione: se osi criticare la digitalizzazione esistente, per quanto nata male e gestita peggio, ti scopri ad essere "contro la digitalizzazione". Strano, perche' ci vivo. Questo e' il metodo col quale si mettono a tacere voci che sono contrarie all'adozione di qualcosa perche' penso che il progetto nasca male, o sotto false premesse. [...] Ci sono due domande principali , che affondano i progetti: 1 Chi paga, ovvero come facciamo a non rimetterci? 2 Come e' strutturata la domanda di questi servizi? [...] In questa fase, se lavorate nel mondo del privato, ci sono DUE cose che NON dovete fare. Due risposte che NON dovete dare. 1 Paga l'ecosistema, o non meglio precisati "privati", che poi fanno i soldi col "valore aggiunto" 2 La domanda che vediamo e' generica perche' non ci sono soluzioni, ma sara' stimolata dalla stessa esistenza del nostro nuovo servizio. Queste due sono le premesse del fallimento catastrofico. Erano le premesse dello SPID. Quindi, purtroppo SPID parte con entrambe le gambe gia' rotte. [...] La verita' e' che si era ideato un servizio che doveva essere riconosciuto legalmente dallo stato, ma non "non doveva costare nulla al cittadino" e "non doveva costare nulla alla PA". Cioe' si proponeva un servizio anagrafico (Identity Provider significa "anagrafe") senza oneri per lo stato, senza oneri per il cittadino, credendo che i privati lo avrebbero utilizzato in massa. Dovete solo avere la pazienza di indagare un attimo per notare che tutti questi providers sono in ballo con lo stato, o come fornitori , o come beneficiari di politiche bancarie, o ex monopolisti di stato, insomma, gente cui poi chiedere , cenando su una terrazza a Roma, "di farti il favorino di badare a questo progettino". Perche' sono quasi certo che la cosa sia andata cosi'? Perche' le aziende normali, quando entrano in un settore digitale (diciamo pure l'identita' digitale) lo fanno con un business plan, un'analisi di mercato, e di una strategia. E se la strategia e' "poi ci guadagnamo coi servizi a valore aggiunto" , insieme al servizio SPID avrebbero creato ANCHE questi servizi VAS: non vai sul mercato senza un prodotto. Se cui vai per guadagnare coi servizi VAS, allora hai anche questi servizi VAS. Li avete visti? [...] https://www.corriere.it/economia/lavoro/23_febbraio_23/spid-rischio-chiusur… In pratica, ci stanno dicendo che lo stato ha fatto partire la baracca pensando che l'avrebbero finanziata i privati vendendo servizi VAS al mercato. I privati che ci sono entrati, invece, pensavano che lo stato li avrebbe sponsorizzati con del credito di imposta, cioe' con soldi pubblici. La vera domanda e': ma si sono mai parlati? Erano presenti quando si disegnava il servizio? Sono stati consultati, almeno? E se no, perche' diavolo sono entrati nel business? La risposta e', ovviamente, solo una: politica. [...] Possiamo riassumere la saga in questo modo: 1 Il 4 luglio 2016 cominciò la sostituzione della carta d'identità cartacea con la CIE,Il progetto pilota risale al 2001. 2 La Zecca dello Stato di Roma fallisce clamorosamente nel consegnare i supporti fisici, o almeno cosi' dicono i comuni. 3 Siccome sta andando tutto bene, e non ci sono code da smaltire (scherzo) nel 2015 si ridefinisce la carta CIE, aggiungendo la modalita' contactless. 4 Arrivano quelli bravi, e nello stesso anno ,28 luglio 2015, con la Determinazione n. 44/2015, sono stati emanati da AgID i dettagli tecnici e giuridici per lo SPID. Il progetto e' stato discusso in Parlamento. 5 Non basandosi sulla Zecca dello Stato e sui suoi nani amanuensi, lo SPID raggiunge i 30 milioni di cittadini perche' la PA lo adotta in massa. 6 Siccome le marmotte del cioccolato non pagano i costi dell'operazione, adesso c'e' un braccio di ferro tra i gestori e lo stato, per circa 50 milioni di spese di supporto che nessuno vuole pagare. 7 I supersboroni supermoderni diggitali dello SPID fanno tifo da stadio contro i Nani di Moria che scolpiscono a mano le CIE col piccone, dentro le caverne. 8 Sia Apple che i produttori di cellulari Android hanno un wallet digitale contactless, e ormai e-sim e' normale. In alcuni stadi ci sono curve di tifosi che inneggiano al CIE , o allo SPID. Cringe intensifies. --8<---------------cut here---------------end--------------->8--- ...e dei supersboroni supermoderni diggitali del wallet integrato nello smathphone o delle eSIM integrate nello smartphone che dire? :-O Saluti, 380° P.S.: e di quelli che non usano le lettere accentate ma l'apostrofo come nel 1985, che dire? (scherzo) :-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>.

2 1

ChatGPT: sicuri che.... non "sappia" nulla?
by Damiano Verzulli Feb. 23, 2023

Feb. 23, 2023

Mi aggancio alla discussione in corso su altri thread sulla natura statistica/probabilistica di ChatGPT e sulle relative implicazioni (non puo' "sapere"; non produce "informazione", quindi non può errare; non ha "conoscenza", etc.). Da tecnico, rilevo che la "qualita'" delle risposte sia fortemente determinata dal dominio della domanda. Se chiedi la biografia di Pinco Pallino da Pennapiedimonte, è un conto. Se chiedi cos'é un "inner join" è un altro conto. Trovo che la cosa sia molto simile al mio carattere personale: se alle superiori mi interrogavano su Manzoni... facevo una fatica bestiale; se mi chiedevano di disegnare una funzione, era un gioco da ragazzi. Dico questo, perché ieri parlavo (mentre sfilavamo per il paese, sul carro di Carnevale) con un amico che mi chiedeva: "Damia'! Ma se ho una relazione N a M fra 2 tabelle SQL, come posso organizzare il database?". Ed io gli dicevo: "Nessun problema. Ti serve una terza tabella, dove infili la relazione fra le due". E ne abbiamo discusso per qualche minuto (attorniati da coriandoli, musica e tanto altro "casino"). Stamattina mi ha scritto, copia-incollandomi l'interazione con ChatGPT: ================= Domanda: "I have a database containing two tables "elementi" and "proprietari" in elementi i have rows that refers to multiple proprietari and vice versa, what can be the best way to use reference keys in this scenarios?" Risposta: "In a scenario where you have a many-to-many relationship between the "elementi" and "proprietari" tables, the best way to use reference keys is to create a third table that acts as a bridge between the two tables. This is commonly referred to as a junction table or an associative table. The junction table should contain foreign keys that reference the primary keys of the "elementi" and "proprietari" tables. This table can also contain additional columns that relate to the relationship between the two tables, such as the date the relationship was established or the nature of the relationship. For example, you could create a junction table called "elementi_proprietari" with columns "id", "elemento_id", and "proprietario_id". The "id" column would be the primary key of the junction table, while "elemento_id" and "proprietario_id" would be foreign keys that reference the "id" column of the "elementi" and "proprietari" tables, respectively. In this way, each row in the junction table represents a relationship between a single "elemento" and a single "proprietario". To represent relationships between multiple "proprietari" and a single "elemento", you would simply add additional rows to the junction table with the same "elemento_id" but different "proprietario_id" values. By using a junction table, you can easily query and update the relationships between "elementi" and "proprietari" using standard SQL commands" ================= Ai non-tecnici che potrebbero non percepire la qualita' della risposta, segnalo che non solo è _PERFETTA_, ma la trovo anche qualitativamente migliore di molte altre spiegazioni che circolano in aria all'inizio dei corsi universitari sui database relazionali. Ora io non voglio mettere in dubbio il fatto che ChatGPT risponda seguendo logiche statistiche o probabilistiche. Voglio solo dire che per _MOLTI_ ambiti (quelli strettamente piu' "tecnici") si fa veramente fatica a dire che il suo "output" segua esclusivamente tale logica o --per tornare ai thread in corso-- che ChatGPT non "sappia" come implementare una relazione N a M nei contesti dei database relazionali. E comunque --qualora si volesse insistere su questo punto [come mi pare alcuni vogliano fare]-- anche se cosi' fosse... tendenzialmente al mio amico non gliene frega un granché [perché la risposta alla sua domanda _TECNICA_, è _PERFETTA_]. Aggiungo --infine-- che pur avendo conseguito una laurea in Scienze dell'Informazione nel lontano 1995, ho scoperto cosa fosse un Gerber-File soltanto nel ~2016. Ed è stato soltanto grazie al mio amico (quello del carro), che ho scoperto l'esistenza della "Saldatura a TIG" un paio di anni fa e che, a dispetto della pressione, a temperatura ambiente non è possibile liquefare il metano (il propano si, invece). Insomma: a 52 anni riesco ancora facilmente ad "apprendere" cose molto tecniche/specifiche.... Allo stesso tempo (continuo) a far fatica a metabolizzare la biografia del Manzoni o gli scritti di Dante. Mi sorge il dubbio, che anche il mio cervello funzioni --da qualche parte-- con logiche statistico/probabilistiche. Un'ultima nota: ho letto piu' volte, in lista, di organizzare forme di "resistenza" a questo fenomeno. Aldila' del fatto che trovo il termine "resistenza" non-brillantissimo, nutro fortissimi dubbi che --perdonatemi la brutalità-- da questa lista, per questo ambito specifico, possano uscire cose piu' concrete degli Archives della lista stessa. Bye, DV -- Damiano Verzulli e-mail: damiano(a)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

9 13

"The EU’s media crackdown: Why redfish is forced to close"
by J.C. DE MARTIN Feb. 23, 2023

Feb. 23, 2023

*The EU’s media crackdown: Why redfish is forced to close** * February 22, 2023 Since 2017, redfish has dedicated itself to reporting from the perspective of people around the world who are oppressed by the capitalist system in the Global North and South and all its features: racism, war, imperialism, patriarchy, ecocide and extreme inequality. Looking back, we are immensely proud of all we have accomplished over the past five years but the reality of developments around redfish this year has forced us to make a hard decision: redfish is closing. The relentless campaign of smears by mainstream media, academia and think tanks; social media censorship and now state-sanctioned harassment have made it impossible for us to continue. Here is what happened to us, how we have been used as one test case in the EU’s weaponization of “disinformation” to shut down critical voices, and why what we’ve been through is only the beginning of a crackdown that will impact everyone, including those who have encouraged and cheered the suppression of our work. The recent Twitter Files have rightly brought a lot of attention to how entrenched the US military and intelligence apparatus is at the executive level of social media giants like Twitter, who control the information we have access to. Completely under the radar, however, the political heart of the EU in close collaboration with allied military institutions and those same social media companies, has developed its own strategy complete with a chilling array of tools to crack down on voices critical of the status quo of Western capitalist imperialism. In EU policy circles the threat is described as “undermining the European project”, and questioning the “democratic legitimacy of the representatives of Member States”. That is, in the eyes of the EU’s decision-makers, voices that publicly question the current political and socio-economic order in the West are no longer entitled to international and EU legal protection which covers the right to freedom of expression. On the contrary, we are to be treated as a threat. [...] continua qui: https://redfish.media/blog/the-eus-media-crackdown-why-redfish-is-forced-to…

2 1

Fwd: [DigHum Lecture Series] Next talks on March 7 Cory Doctorow “Seize the Means of Computation” & March 9 Gary Marcus "On ChatGPT"
by Antonio Vetro' Feb. 23, 2023

Feb. 23, 2023

Segnalo che i prossimi due incontri della serie Digital Humanism vertono su argomenti molto discussi nella lista in questo periodo. Dettagli sotto, estrapolati dalla loro mailing list. Antonio *** Tuesday, March 07, 2023 at 5:00 p.m. (17:00) Central European Time (UTC+1) Topic: “Seize the Means of Computation” (scroll down for abstract and CV) Speaker: Cory Doctorow (craphound.com) Moderator: Allison Stanger (Middlebury College, USA) *** Thursday, March 09, 2023 at 5:00 p.m. (17:00) Central European Time (UTC+1) Topic: “On ChatGPT” (scroll down for abstract and CV) Speaker: Gary Marcus (garymarcus.com) Moderator: Helga Nowotny (Chair of the ERA Council Forum Austria and Former President of the ERC) *** To participate in the talks via Zoom go to: https://tuwien.zoom.us/j/96389928143?pwd=UU5YRkNuRmdoWHV4MFBwMWRCcUErdz09<http://xuxss.mjt.lu/lnk/AW0AAA91AZcAAAAhADgAAAHQX18AAAAARcEAADJzABG7CQBj92b…> (Password: 0dzqxqiy) Both talks will be live streamed and recorded on our YouTube channel: https://www.youtube.com/digitalhumanism<http://xuxss.mjt.lu/lnk/AW0AAA91AZcAAAAhADgAAAHQX18AAAAARcEAADJzABG7CQBj92b…> For further announcements and information about the speakers in the Lecture Series, see https://dighum.ec.tuwien.ac.at/news-events<http://xuxss.mjt.lu/lnk/AW0AAA91AZcAAAAhADgAAAHQX18AAAAARcEAADJzABG7CQBj92b…>. Please note that you can access the slides and recordings of our past events via that link. ----- ABSTRACT “Seize the Means of Computation”: The tech giants claim that they have to lock down their devices to defend you, but they keep failing to do so. The most science fictional question isn’t what a technology does, it’s who it does it for and to. A free and fair future starts with technology that is under its users’ control. We need a new Luddite movement, one that seizes control of the machines that try to seize control of us. Short Bio of Cory Doctorow: Cory Doctorow (craphound.com) is a science fiction author, activist, and journalist. His latest book is CHOKEPOINT CAPITALISM (with Rebecca Giblin) nonfiction about creative labor markets and monopoly. His latest novel is ATTACK SURFACE, a standalone adult sequel to LITTLE BROTHER. He is also the author HOW TO DESTROY SURVEILLANCE CAPITALISM, nonfiction about conspiracies and monopolies; and of RADICALIZED and WALKAWAY, science fiction for adults, a YA graphic novel called IN REAL LIFE; and young adult novels like HOMELAND, PIRATE CINEMA and LITTLE BROTHER. His first picture book was POESY THE MONSTER SLAYER (Aug 2020). In 2023/4, Tor Books will publish two more science fiction novels for adults: RED TEAM BLUES and THE LOST CAUSE; and Verso will publish THE INTERNET CON, a nonfiction book about monopoly and radical interoperability. He maintains a daily blog at Pluralistic.net. He works for the Electronic Frontier Foundation, is a MIT Media Lab Research Affiliate, is a Visiting Professor of Computer Science at Open University, a Visiting Professor of Practice at the University of North Carolina’s School of Library and Information Science and co-founded the UK Open Rights Group. Born in Toronto, Canada, he now lives in Los Angeles. In 2020, he was inducted into the Canadian Science Fiction and Fantasy Hall of Fame. ABSTRACT "On ChatGPT": Gary Marcus is a prominent public voice on the developement of AI, see his recent NY Times interview <http://xuxss.mjt.lu/lnk/AW0AAA91AZcAAAAhADgAAAHQX18AAAAARcEAADJzABG7CQBj92b…> or his comments on his blog<http://xuxss.mjt.lu/lnk/AW0AAA91AZcAAAAhADgAAAHQX18AAAAARcEAADJzABG7CQBj92b…>. In the DigHum Lecture he will present his latest take on ChatGPT. Short Bio of Gary Marcus: Gary Marcus is a leading voice in artificial intelligence. He is a scientist, best-selling author, and serial entrepreneur (Founder of Robust.AI and Geometric.AI, acquired by Uber). He is well-known for his challenges to contemporary AI, anticipating many of the current limitations decades in advance, and for his research in human language development and cognitive neuroscience. An Emeritus Professor of Psychology and Neural Science at NYU, he is the author of five books, including, The Algebraic Mind, Kluge, The Birth of the Mind, and the New York Times Bestseller Guitar Zero. He has often contributed to The New Yorker, Wired, and The New York Times. His most recent book, Rebooting AI, with Ernest Davis, is one of Forbes’s 7 Must Read Books in AI.

1 0

A proposito del pluridiscusso Cyber-Resilience-ACT: una lista delle critiche
by Damiano Verzulli Feb. 23, 2023

Feb. 23, 2023

Piu' o meno per caso, mi è appena passato sotto agli occhi questo: https://blog.opensource.org/the-ultimate-list-of-reactions-to-the-cyber-res… "The European Commission’s proposed Cyber Resilience Act (CRA) as drafted may harm Open Source, and perhaps all other non-industrial software. There were 131 responses to the proposed text that the Commission has sent to the Parliament, including one from the Open Source Initiative. Of those, 18 responses – representing a significant proportion of Europe’s software industry – shared OSI’s concerns to some degree. Here are some sample points from the responses" ...e poi seguono i testi (e i link) delle 18 entita' che condividono (e mettono nero-su-bianco) le perplessita' di OSI: * Open Source Initiative (OSI) * Open Forum Europe (OFE — with OSI, Eclipse, APELL, CNLL, OSBA) * The Document Foundation (LibreOffice) * Vrijschrift.org * NLNet Labs (with CZ.NIC, ISC, NetDEF) * Developers Alliance * RIPE NCC * ITI – Information Technology Industry Council * DIGITALEUROPE * Japan Business Council in Europe (JBCE) * Bitkom * Eco * OpenSSF (Linux Foundation) * OpenXchange * GitHub * Huawei * Microsoft * Sonatype Magari, assemblare qualcosa di simile, rispetto ai discorsi di stamane (su IA/ChatGPT e dintorni), dove nei punti elenco ci sono (almeno) i vari contributori di questa lista.... potrebbe essere qualcosa di interessante Bye, DV -- 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

1 0