While testing a web application as part of a bug bounty program, I uncovered a critical RCE vulnerability by chaining directory traversal with a subtle CSV parsing abuse.
If your default comment sort isn’t “hot”, the when you look at the comments on a post it will be sorted by “hot”, your comment sort will only be applied to a post once you refresh that post.
C’è qualcosa di profondamente inquietante nel constatare che un videogioco che hai regolarmente comprato, pagato, installato e magari amato… possa semplicemente smettere di esistere. Non per un guasto alla tua console. Non per tua scelta.
An internal Microsoft memo has leaked. It was written by Julia Liuson, president of the Developer Division at Microsoft and GitHub. The memo tells managers to evaluate employees based on how much t…
FWIW, I work in a field that is mostly related to law and accounting. Unlike with coding, there are no simple "tests" to try out whether an AI's answer is correct or not. Of course, you could try these out in court, but this is not something I would recommend (lol).
In my experience, chatbots such as Copilot are less than useless in a context like ours. For more complex and unique questions (which is most of the questions we are dealing with everyday), it simply makes up smart-sounding BS (including a lot of nonexistent laws etc.). In the rare cases where a clear answer is already available in the legal commentaries, we want to quote it verbatim from the most reputable source, just to be on the safe side. We don't want an LLM to rephrase it, hide its sources and possibly introduce new errors. We don't need "plausible deniability" regarding plagiarism or anything like this.
Yet, we are being pushed to "embrace AI" as well, we are being told we need to "learn to prompt" etc. This is frustrating. My biggest fear isn't to be replaced by an LLM, not even by someone who is a "prompting genius" or whatever. My biggest fear is to be replaced by a person who pretends that the AI's output is smart (rather than filled with potentially hazardous legal errors), because in some workplaces, this is what's expected, apparently.
I work in a field that is mostly related to law and accounting... My biggest fear is to be replaced by a person who pretends that the AI’s output is smart
Aaaaaah. I know this person. They're an accountant. They recently learned about AI. They're starting to use it more at work. They're not technical. I told them about hallucinations. They said the AI rarely wrong. When he's not 100% convinced, he says he asks the AI to cite the source.... 🤦 I told him it can hallucinate the source! ... And then we went back to "it's rarely wrong though."
And then we went back to “it’s rarely wrong though.”
I am often wondering whether the people who claim that LLMs are "rarely wrong" have access to an entirely different chatbot somehow. The chatbots I tried were rarely ever correct about anything except the most basic questions (to which the answers could be found everywhere on the internet).
I'm not a programmer myself, but for some reason, I got the chatbot to fail even in that area. I took a perfectly fine JSON file, removed one semicolon on purpose and then asked the chatbot to fix it. The chatbot came up with a number of things that were supposedly "wrong" with it. Not one word about the missing semicolon, though.
I wonder how many people either never ask the chatbots any tricky questions (with verifiable answers) or, alternatively, never bother to verify the chatbots' output at all.
AI fans are people who literally cannot tell good from bad. They cannot see the defects that are obvious to everyone else. They do not believe there is such a thing as quality, they think it's a scam. When you claim you can tell good from bad, they think you're lying.
The "right" pronunciation depends on what you take for reference. If you're treating the word as an English one, the woman is right - English doesn't allow this sort of initial cluster, and spelling-wise it's well-established that some initial consonants are mute (see e.g. "knife").
And, if treating the word as a Koine Greek one... odds are both are butchering the word so much that [pə] wouldn't save the day. Even the man is likely pronouncing it as [pə.'tɒl.ə.mi]; the Greek word would be more like [pto.le.'mɛ:.os], or [pto.le.'maɪ̯.os] with a really conservative pronunciation.
As it happens, "Ptolemaois" is how the name is written in at least German, Swedish and Finnish, so speakers of those languages (Swedish and Finnish myself) likely pronounce it most correctly?
Never really understood why English insists of weirdly dropping the final bits of Greek and Latin names ("Plutarch" vs "Plutarkhos", "Justinian" vs "Justinianus" etc)
If by "most correctly", you mean "the closest to what Koine Greek would do", then yes. Note however that each language will impose restrictions on the allowed sounds and sequences of; for example Finnish won't use [ä] like Ancient Greek would, simply because the sound isn't there in Finnish (it adapts it to an [ɑ]).
Also note the word itself can be pronounced multiple ways even in Koine Greek. For example the ⟨αῖ⟩ diphthong can be read as either [äɪ̯] (as in English "by") or as [ɛ:] (as in English air); as far as I'm aware this sound change happened in early Koine Greek times.
Never really understood why English insists of weirdly dropping the final bits of Greek and Latin names (“Plutarch” vs “Plutarkhos”, “Justinian” vs “Justinianus” etc)
Short explanation: English does it because it's what French does. And French does it because of its history as a Latin descendant.
Long explanation:
Since French is a Romance language, it's the result of a Latin dialect undergoing a bunch of sound changes. Those sound changes affected all words inherited from Latin. For example capus/capum¹ → chef, bonus/bonum → bon, Romanus/Romanum → Romain (yup, it applies to personal names!) ille → le, so goes on.
However, Latin is a prestige language in Europe. So even if French is a Latin descendant, it kept reborrowing words from Latin. And because of the above, French started changing those loanwords in a specific way, that kind of mimics part of its own evolution.
In other words: French developed a convention on how to handle Latin borrowings². And part of that convention is to sub/remove the endings. Other Romance languages do something similar³.
What I said applies to the Latin names. Now, the Greek names go one step deeper: Latin itself borrowed Greek words left and right, adapting them into Latin. Some would be eventually inherited by French. So the convention on how to handle Latin names in French also handles Greek names: "Latinise them first, then pretend they're Latin words."
Then you get English. Most of that Classical knowledge entered English through French, so English borrowed that convention of adapting Latin words too. Eventually developing its own convention on how to do it, that looks kind of similar to the one French used back then. And some names were subjected to local sound changes, and just like the Romance languages English messes a fair bit with word endings. And the vowels, too (Great Vowel Shift).
In contrast, German also treats Latin as a prestige language. But since it's neither a Romance language nor borrowing the convention from one, it's getting the names straight from Latin, and modifying them a bit less⁴. That includes keeping the nominative endings of the words.
NOTES:
I'm listing words by their Latin nominative and accusative. The nominative is the form likely to be borrowed; however, French and the other Romance languages inherited the accusative.
This can be seen by the Modern French renditions of those names: Ptolémée, Justinien, Plutarque.
For reference, look at the Italian versions of those names: Tolomeo, Giustiniano, Plutarco. Parts of the ending are still there, unlike in French, but the ending -s/-m is gone.
It still does change them, mind you. After a word is borrowed into a language, it's subjected to the sound changes of that language; plus spelling plays a huge role, and even in non-Romance languages there are minor conventions on how you're "supposed" to handle Latin names. Cue to German spelling "Justinianus" instead of "IVSTINIANVS" or "Iustinianus".
A new ProPublica data analysis adds to the mounting evidence that abortion bans have made the common experience of first-trimester miscarriage far more dangerous.
Or: What are the best microSD cards you can get for under $15 in 2024 2025? If you just want to see my top picks, skip to the Overall Picks section. If you just want to see the raw data, click here. If you want to read in more detail about how I...
I think the 17 year old sees them because they've got their age range at like 18-20, loads of underage people make Tinder accounts and put their real age in the bio. And the 29 year old would then see a lot fewer men. That's what I'm guessing the comment you replied to was about.
but that would be a crime! instead, we'll just ask the companies to do the right thing, surely the only reason they're not already doing that is because they're unaware
I assume game studios are currently extremely well secured on par with film production considering how much money goes into it. Back in the day, maybe not so much. But then you might also be limited by the capabilities of portable data storage. Kind of harder to smuggle out a dozen floppy disks than one thumb drive. Plus, they literally get the FBI to investigate if they notice something gets stolen, so it's really not a simple thing to just walk out of work one day with the source code of the game you've been working on for years.
2 guys looking out at hell and one says 'it is even worse than I imagined' there are people all over the place saying things like "must be nice staying home all day", "yoga cured my cousin", "you don't look sick", "just be positive", "we are all tired" Artist is Glenn McCoy
Divulgazione e social: quando la competenza diventa un bersaglio
Non sono un frequentatore abituale di X (ex Twitter), non è il mio habitat naturale. Preferisco di gran lunga il fediverso, e in particolare Mastodon, dove il dialogo è più disteso, senza algoritmi e più rispettoso delle competenze. Continuo però a mantenere un account su X, principalmente per seguire voci che considero autorevoli e preziose.
Tra queste c’è un geologo, divulgatore scientifico e saggista italiano, noto anche per la sua presenza televisiva, che non si limita a spiegare: provoca, stimola, prende posizione. Lo seguo da anni, e apprezzo il suo coraggio nel dire cose scomode, specie quando si tratta di ambiente, pseudoscienza o gestione del territorio.
È proprio questa chiarezza che divide: c’è chi lo apprezza per la capacità di semplificare temi complessi e chi, al contrario, lo accusa di essere arrogante o ideologico.
Sui social tradizionali non esiste più un ambito di competenza: tutto si riduce a opinioni. Un geologo e divulgatore scientifico con trent’anni di esperienza viene contestato da profili “qualunque” che, senza alcuna preparazione, si sentono legittimati a correggerlo o ridicolizzarlo.
Sotto ogni suo post si accumulano commenti sprezzanti, battute, meme e accuse. C’è chi nega il cambiamento climatico, chi lo accusa di essere al servizio di qualche lobby o di voler fare solo propaganda. Il punto non è confutare con dati certi, ma sminuire, ironizzare, creare rumore. Sarebbe interessante vedere quegli stessi, così baldanzosi dietro la tastiera, trovarsi davanti alla lavagna a confutare i loro commenti, a voce alta, penna in mano.
Le piattaforme come X premiano la viralità, non l’autorevolezza. Più un contenuto genera reazioni, anche negative, più viene spinto in alto nei feed.
Il risvolto della medaglia? L’esperto, suo malgrado, può diventare una figura popolare proprio grazie alle polemiche. L’algoritmo trasforma il conflitto in visibilità, e in certi casi fa conoscere il divulgatore anche a chi altrimenti non l’avrebbe mai incontrato. Ma si tratta di una notorietà probabilmente distorta, basata più sull’arroganza e la cattiveria gratuita di chi attacca che sull’interesse reale per i contenuti.
La visibilità diventa così un’arma a doppio taglio: porta l’informazione a un pubblico maggiore, ma lo fa in un contesto saturo di ostilità.
Negli ultimi anni si è diffusa una retorica secondo cui ogni opinione ha lo stesso valore, a prescindere da chi la esprime. L’idea che “uno vale uno” si è trasformata, sui social, in un attacco sistematico verso chi possiede conoscenze reali, lavora con dati verificati o applica il metodo scientifico. L’idea che “uno vale uno” sui social tradizionali si è trasformata. L’espressione “uno vale uno” è spesso usata per affermare che ogni persona ha lo stesso diritto di parola. Questo è sacrosanto in una democrazia. Il problema nasce quando questo principio viene frainteso e usato per sostenere che tutte le opinioni si equivalgano, anche quando una è supportata da anni di studio e dati verificati e l’altra è solo un’impressione personale o una teoria campata in aria.
Chi rappresenta competenza viene spesso percepito come arrogante o distante, e non di rado diventa bersaglio di sfiducia o rabbia. Non serve argomentare, basta insinuare dubbi, sbeffeggiare, trasformare ogni affermazione in un’occasione per una battuta a effetto.
Esiste un’alternativa?
In ambienti come il Fediverso, ad esempio Mastodon, dove la conversazione si sviluppa senza intervento degli algoritmi i toni sono meno tossici, la comunicazione meno aggressiva, il rispetto per la competenza più presente e la visibilità non dipende da quanto rumore si riesce a fare.
Questo non significa che tutti siano d’accordo, ma il dissenso tende a manifestarsi in modo più civile e costruttivo. Per chi, come l’esperto che seguo, cerca un dialogo autentico potrebbe valere la pena esplorare questi spazi nel Fediverso.
I social tradizionali ormai non sono più solo luoghi di conversazione ma ambienti tossici dove il dibattito diventa scontro.
Forse è il momento di chiederci: vogliamo davvero continuare ad affidare la qualità del dibattito pubblico a piattaforme che premiano il rumore anziché il merito?
Three years [ago], the governor of St. Petersburg signed a sister-city agreement with the occupying authorities of Mariupol, the Ukrainian port city that was razed to the ground in a devastating Russian siege just weeks beforehand.
“Since then, St. Petersburg has hosted children from Mariupol for every camp session — both in summer and winter,” said Governor Alexander Beglov.
This summer, Russian authorities are organizing five three-week camp sessions for children from the occupied city. Each session is led by child psychologists, St. Petersburg schoolteachers and camp counselors who recently graduated from teacher training college.
More than 2,000 schoolchildren from Mariupol in total are expected to attend camps in St. Petersburg this year.
Initially, Russian authorities billed these summer programs as health and wellness retreats for children who had lived under Russian shelling.
But from the very first sessions, children were also taught to develop respect and love for the country that seized their home city.
[...]
Today Ukraine has confirmed the deportation of 19,546 children from occupied Ukrainian territories to Russia, though experts say the real number is likely much higher.
In March 2023, the International Criminal Court issued arrest warrants for President Vladimir Putin and his children’s rights commissioner Maria Lvova-Belova in connection with these deportations.
Ukrainian experts say Russia is deliberately stripping these children of their Ukrainian identity and raising them to become Russians, turning minors into a new generation loyal to the Kremlin.
The educational program at Camp Druzhnykh lists goals that include fostering a national — that is, Russian — identity among the children.
[...]
The camp also organizes a career fair where children can learn about the job market in Russia. In June, it featured a police college that accepts students as early as ninth grade. Students from the college spoke to the children about the ceremonial police oath and showed them how to take fingerprints.
[...]
Now in high school, Masha [a girl form Mariupol, not her real name] quietly dreams of moving to St. Petersburg for university. But when she talks about the future, there is a sadness in her voice [...] “I used to think living in Russia was easy. But then my mom tried to get a job at Pyaterochka [a discount supermarket chain], and the salary was under 20,000 rubles (less than $253) — while the country’s minimum subsistence level is 17,000 ($215). That’s when I realized life in Russia is hard. You don’t live — you survive.”
RSS feeds are the way to go to keep in touch with events, but only when you want to and in a not addictive manner. Also, my philosophy is that some news is really really important only if somebody from my inner circle personally tells me something about it. Otherwise is just noise from media
Giugno 2025 si è chiuso con temperature eccezionalmente elevate in Toscana: oltre 40 gradi a Larciano e Massa, 38.4 a Cascina, dove per ben 20 giorni la colonnina di mercurio ha superato i 32 gradi. Nessuna pioggia, e un deficit di 38 millimetri rispetto alla media. Ma il dato più preoccupante non è solo il caldo in sé, bensì la frequenza con cui questi eventi si ripetono. È il segnale del cambiamento climatico in corso, e il rischio è quello dell’assuefazione: abituarci a condizioni che un tempo erano eccezionali. Scopri l’analisi completa su [nome del sito] e approfondisci cosa sta davvero cambiando nel nostro clima.
Il mese di giugno 2025 si è chiuso con un’ondata di calore intensa e persistente che ha investito gran parte della Toscana, registrando temperature decisamente sopra la media. I dati
Myself and some other users have been getting confused why we couldn't crosspost text posts.
Eg. I can't crosspost this https://piefed.social/post/993050 to https://piefed.
David Seymour is an evil individual and we as a society should be ashamed that we have given him this much power and that the media treat him with any degree of respect.
Profitez des vidéos et de la musique que vous aimez, mettez en ligne des contenus originaux, et partagez-les avec vos amis, vos proches et le monde entier.
TrueBlue Exhibits offers comprehensive solutions for trade show exhibits, branded environments, and corporate events, as well as LED video wall rentals and union labor services, all in one convenient location.
Palestinian officials say Israeli airstrikes killed more than 60 people, including at a cafe in northern Gaza and outside a food distribution site in southern Gaza. The violence comes as President Trump is making a push this week for a ceasefire.
Martedì 8 luglio torniamo con il Logout di TWC Roma, il ritrovo per tech workers che vogliono incontrarsi dopo lavoro: un'occasione per socializzare, conoscersi, parlare del nostro lavoro e come organizzarci nei prossimi mesi!
Ci vediamo martedì 8 luglio, alle 18.30, alla casa del parco della Caffarella
Quante volte ci si chiede se un musicista ormai datato della scena artistica mondiale, con decine e decine di album pubblicati, abbia ancora qualcosa di nuovo da farci sentire? ... Leggi e ascolta...
Quante volte ci si chiede se un musicista ormai datato della scena artistica mondiale, con decine e decine di album pubblicati, abbia ancora qualcosa di nuovo da farci sentire? E' molto probabile che, se il musicista non appartiene alla nostra sfera di preferenza, lo si liquidi subito, a volte ancor prima di ascoltarlo, con un bel “niente di nuovo”, “album inutile”, “ormai finito” ecc. ecc.; se invece è un nostro beniamino o ancor meglio apparteniamo alla sfera dei suoi fans incalliti, è molto probabile che il nostro giudizio sia “oscurato” dal classico velo affettivo che, per carità non è “pietoso”, ma senz'altro poco obbiettivo. Ecco, tutta questa premessa per arrivare a dire che, personalmente, pur appartenendo alla seconda sfera, quella dei fan incalliti e non da tempi recenti, serenamente affermo che Wrecking Ball è un buon album e il nostro sessantatreenne “The Boss” riesce a dirci ancora molto!... artesuono.blogspot.com/2014/07…
HedyL
in reply to David Gerard • • •FWIW, I work in a field that is mostly related to law and accounting. Unlike with coding, there are no simple "tests" to try out whether an AI's answer is correct or not. Of course, you could try these out in court, but this is not something I would recommend (lol).
In my experience, chatbots such as Copilot are less than useless in a context like ours. For more complex and unique questions (which is most of the questions we are dealing with everyday), it simply makes up smart-sounding BS (including a lot of nonexistent laws etc.). In the rare cases where a clear answer is already available in the legal commentaries, we want to quote it verbatim from the most reputable source, just to be on the safe side. We don't want an LLM to rephrase it, hide its sources and possibly introduce new errors. We don't need "plausible deniability" regarding plagiarism or anything like this.
Yet, we are being pushed to "embrace AI" as well, we are being told we need to "learn to prompt" etc. This is frustrating. My biggest fear isn't to be replaced by an LLM, not even by someone who is a "prompting genius" or whatever. My biggest fear is to be replaced by a person who pretends that the AI's output is smart (rather than filled with potentially hazardous legal errors), because in some workplaces, this is what's expected, apparently.
paequ2
in reply to HedyL • • •Aaaaaah. I know this person. They're an accountant. They recently learned about AI. They're starting to use it more at work. They're not technical. I told them about hallucinations. They said the AI rarely wrong. When he's not 100% convinced, he says he asks the AI to cite the source.... 🤦 I told him it can hallucinate the source! ... And then we went back to "it's rarely wrong though."
HedyL
in reply to paequ2 • • •I am often wondering whether the people who claim that LLMs are "rarely wrong" have access to an entirely different chatbot somehow. The chatbots I tried were rarely ever correct about anything except the most basic questions (to which the answers could be found everywhere on the internet).
I'm not a programmer myself, but for some reason, I got the chatbot to fail even in that area. I took a perfectly fine JSON file, removed one semicolon on purpose and then asked the chatbot to fix it. The chatbot came up with a number of things that were supposedly "wrong" with it. Not one word about the missing semicolon, though.
I wonder how many people either never ask the chatbots any tricky questions (with verifiable answers) or, alternatively, never bother to verify the chatbots' output at all.
David Gerard
in reply to HedyL • • •sturger
in reply to David Gerard • • •In other words, AIs are BS automated BS artists... being promoted breathlessly by BS artists.
Honytawk
in reply to sturger • • •LLMs have their flaws, but to claim they are wrong 70% of the time is just hate train bullshit.
Sounds like you base this info on models like GPT3. Have you tried any newer model?
Rocco Prestia's Basilisk
in reply to Honytawk • • •