←
Dave Stokes21,530 posts
Dave Stokes avatar

Dave Stokes

@stoker
◷ Joined January 2008▧ 1,203 Photos and videos
𝕐o̴g̴ avatar
⟳ Dave Stokes retweeted
𝕐o̴g̴@Yoda4ever·9h

Working from home..🐕🐑🐑😅

0:09
OurSQL Foundation avatar
⟳ Dave Stokes retweeted
OurSQL Foundation@OurSQLFdn·5h

Can AI make old-school DBAs more important? @stoker on why he left Oracle, the cost of “free” Postgres, and what companies lose when they replace juniors with AI. Read or watch the interview: oursqlfoundation.org/blog/da… #MySQL #OpenSource #OurSQLFoundation

DBeaver avatar
⟳ Dave Stokes retweeted
DBeaver@dbeaver_news·8h

You can set your own rules for the SQL queries that AI Chat generates. Use custom prompts for keyword case, JOIN style, formatting, or any other conventions you or your team follow. Custom prompts are available in all DBeaver desktop editions.

0:57
Gabe avatar
⟳ Dave Stokes retweeted
Gabe@fishjar·Nov 16

很多开发团队为了追求极致精简的 Docker 镜像,习惯把基础镜像一律换成仅有 5MB 的 Alpine Linux,然而这往往是生产环境神秘崩溃噩梦的开端。 代码在本地 Ubuntu 开发机上跑得稳如磐石,一旦打包成 Alpine 容器上线,高并发下服务就会频繁发生无预警的 SIGSEGV 段错误与直接退出。 很多人第一反应是内存泄漏,但真正的元凶是底层 C 运行时库的隐藏硬限制。标准的 Ubuntu/Debian 采用 glibc,通过 pthread_create 创建的每个新线程,默认会慷慨分配 8 MB 的栈空间(Stack Size)。 而 Alpine 出于极致轻量化的目标,采用了小巧的 musl libc,其线程默认栈空间被暴力砍到了只有区区 128 KB。 当业务代码调用了嵌套深层的 JSON 序列化库、执行了复杂的递归 AST 解析,或者在局部函数内声明了较大的缓冲区结构体时,128 KB 的内存栈瞬间发生溢出击穿保护页,操作系统内核毫无商量余地直接下发 SIGSEGV 强杀进程。在轻量化与工业容错之间,盲目削减基础镜像往往会付出沉重代价。

Gwen (Chen) Shapira avatar
⟳ Dave Stokes retweeted
Gwen (Chen) Shapira@gwenshap·Nov 17

I noticed that folks with kids are better at using AI agents. Probably because they have lots of experience in instructing someone intelligent who looks for loopholes in order to claim that a task was accomplished.

DBeaver avatar
⟳ Dave Stokes retweeted
DBeaver@dbeaver_news·Nov 16

🧩 If your data is spread across two or more databases, you can combine them with the Multi-Source Database driver and query them as one database. It works with any relational databases you've connected in DBeaver. You can also ask AI Chat to write the cross-database query for you. This driver is available in DBeaver Ultimate and DBeaver Team Edition starting with version 26.2.

0:54
DBeaver avatar
⟳ Dave Stokes retweeted
DBeaver@dbeaver_news·Nov 16

We continue to support and develop the Community version, but also offer a quite broad range of PRO tools 🦫 Plus, DBeaver Community is not just free, it's open-source

Mathelirium avatar
⟳ Dave Stokes retweeted
Mathelirium@mathelirium·Nov 15

Its unbelievable that Gradient Descent is so remarkably simple and yet powers almost all of modern AI.

This is how a machine finds the best line through data: it walks downhill. The cost is J(θ₀, θ₁) = 1/(2m) Σ (h(xᵢ) − yᵢ)², with h(x) = θ₀ + θ₁x. Each step does θⱼ := θⱼ − (α/m) Σ (h(xᵢ) − yᵢ)xᵢ. Two runs start at the same point on 14 data points. α = 0.6 zig-zags across the narrow valley of J, α = 0.3 slides along it. The residuals shrink with every step. Every frame is real gradient descent; the bell rings when the gradient is a thousandth of its start.

0:23
Marcus Hutchins avatar
⟳ Dave Stokes retweeted
Marcus Hutchins@MalwareTechBlog·Nov 15

FFMPEG had one of the most valid crash outs in infosec history IMO. There's a huge elephant in the room when it comes to the fact that cybersecurity is extremely unbalanced. You have like 1,000 people pointing at problems for every 1 person fixing them. Vulnerability research is extremely important, but it's become somewhat of a marketing stunt. First with fuzzers, and now with LLMs. It's a lot of "look what my new toy can do" rather than earnest attempts at actually fixing security problem. What Google did to open source developers was especially egregious. They're a $3.6 trillion dollar company, who spent billions upon billions of dollars building a bot that points at problems really really fast, then they just dumped them all on a bunch of unpaid open source project maintainers to fix. Then because of Google's 90 day disclosure policy, the maintainers have 90 days to fix the issues. After 90 days, Google goes public with the issues, making their project look bad if the vulnerability is yet to be fixed, and also potentially harming users if it gets exploited. The 90 day policy was intended to give developers a reasonable amount of time to fix issues, while also applying some pressure with the threat of disclosure. Otherwise, you do get a bunch of goober ass vendors that will just ignore all security issues reported to them, then act completely surprised when their software gets hacked. But when you're a small team, non-profit, or a solo developer, and you have a trillion dollar company's AI just blasting you with infinite problems to fix (each coming with a 90 countdown & a threat), that becomes extremely untenable. They're essentially inadvertently blackmailing developers into becoming unpaid interns. Then on top of that, Google profits unimaginable amounts of money from the use of open source code. They use FFMPEG's code extensively across several highly-profitable products, meanwhile FFMPEG's developers work for free. There's no reason Google couldn't have donated money, engineers, AI tokens, or something towards fixing the problems. They can more than afford it. I don't think Google has any malicious intent at all, it's just that the whole incentive structure behind vulnerability research is completely whack. All the glory goes to people who find issues, not those who fix them, so there's not a whole lot of interest in helping with fixes. Some vulnerability researchers will submit fixes with their reports, but most don't. Honestly, in 2026 cybersecurity basically just feels like the equivalent of 50,000 people following the housecleaner around yelling "you missed a spot". Abysmally little effort has gone into anything outside of pointing out vulnerabilities.

FFMPEG was right all along and people gave them too much shit for it

Sunny Bains @TiDB avatar
⟳ Dave Stokes retweeted
Sunny Bains @TiDB@sunbains·Nov 13

As I suspected current Sysbench is very inefficient. I rewrote it C++ with io_uring and the numbers make much more sense now:

Post attachment
Ben Dicken avatar
⟳ Dave Stokes retweeted
Ben Dicken@BenjDicken·Nov 14

Running a Big™ database has a lot more fun (hard) engineering challenges than small ones. This blog has got to be the best resource on the internet for understanding this. Some Saturday reading, pairs nicely with a cappuccino fyi. planetscale.com/blog/the-lif…

Crunchy Data avatar
⟳ Dave Stokes retweeted
Crunchy Data@crunchydata·Nov 12

Today is Postgres 14's EOL. When launched just over 5 years ago, it added / improved: • SEARCH and CYCLE clauses for recursive CTEs • Multirange types, so a set of ranges fits in one value • date_bin(), which buckets timestamps into fixed intervals • Subscripting for jsonb: data['user']['id'] • Pipeline mode in libpq, which sends several queries before waiting on results • pg_stat_wal, pg_stat_progress_copy and pg_stat_replication_slots views • pg_stat_statements tracking top-level and nested statements separately • VACUUM that gets more aggressive as a database nears transaction ID wraparound • B-tree index updates that reduce bloat

DBeaver avatar
⟳ Dave Stokes retweeted
DBeaver@dbeaver_news·Nov 12

⌨️ Shortcut of the week: maximize the results panel. To see more rows of a large result set, press Ctrl+Shift+T in the SQL Editor. The results panel fills the editor area. Pressing the same shortcut again restores the normal layout.

0:13
Nancy Pelosi Stock Tracker ♟ avatar
⟳ Dave Stokes retweeted
Nancy Pelosi Stock Tracker ♟@pelositracker·Nov 11

You just can’t make this up Rep. Ro Khanna sold Nvidia $NVDA perfectly at the top One month later, CEO Jensen Huang admitted to an affair with a suspected Chinese spy $NVDA is down 26% since Khanna sits on the House committee that investigates China’s attempts to steal US tech To be fair, he's said his trades are run by a 3rd-party But he's also stated “Members of Congress should not be trading stocks”

Post attachment
John Bumstead avatar
⟳ Dave Stokes retweeted
John Bumstead@RDKLInc·Nov 10

Decided to finally see what this AI thing is all about

Post attachment