Can someone give a strong rationale for a separate built-in class? Because "it prevents any unintended modifications" is a bit weak.
If you have fixed keys, a frozen dataclass will do. If you don't, you can always start with a normal dict d, then store tuple(sorted(d.items())) to have immutability and efficient lookups (binary search), then throw away d.
> variable names must not be longer than 6 characters
My memory might be lapsing here, but I don't think MISRA has such a rule.
C89/C90 states that _external_ identifiers only matter up to their first 6 characters [1], while MISRA specifies uniqueness up to the first 31 characters [2].
To make a more specific example, if you malloc()/free() within a loop, it's unlikely that the compiler will fix that for you. However, moving those calls outside of the loop (plus maybe add some realloc()s within, only if needed) is probably going to perform better.
That is something that can be easily found and usually fixed with trivial profiling. I'm more talking about data locality instead of pointer chasing. Once you set up a pointer-chasing data infrastructure changing that means rewriting most of your application.
I guess the crew has to stay pretty close to the end of the hose or it becomes hard to time the... flow... correctly. Likely, they still had to process the frames anyway to make the... flow... look like it comes out of Sterling's mouth, not from the side of his face, so it was basically no extra cost.
Not my kids, but I also passed my Networking course thanks to his guide (early 10s), and I used it as reference material for teaching about sockets in an Operating Systems course this year.
Used his book to learn networking in the late 10s! It's a timeless book at this point. Using the C/Unix socket APIs as the foundation is fantastic. He dives into code and actual network function so well for such a quick read.
No. CERN (aka, Tim Berners-Lee) came up with HTTP and HTML, but he built on top of TCP/IP and DNS.
> using different protocols besides TCP
QUIC is going in that direction, by running on top of UDP.
---
In any case, without replacing the protocol stack we already have different webs, thanks to the walled-garden nature of modern social networking platforms. Linking to an Instagram reel or a Tiktok video is a pain in the neck; if you do not have an account there's a good chance you won't be able to see the content anyway. X is going in a similar direction. This (as well as their non-textual nature, of course) makes them hard to crawl and index for search engines. Fragmentation and niches inevitably ensue.
However - if Java was Oracle to begin with (and as successful in the mid-90s) then might have done some marketing for the Java+Oracle mix.
Some people (ie Managers) if they decide on using Microsoft products will likely "encourage" the use of C# and .NET. -- That is an example of C# + Sql Server.
Most importantly, unless they explicitly use "good" models trained on ethically-sourced data, their reliance on AI is fundamentally at odds with their mission of a "computing movement rooted in human dignity".
I have a pet theory this is due to a wet pod tray: the pod's film partially melts and sticks to the tray before it's released. I made a habit of wiping the tray dry with a piece of paper before loading the pod and this stopped happening to me.
If you have fixed keys, a frozen dataclass will do. If you don't, you can always start with a normal dict d, then store tuple(sorted(d.items())) to have immutability and efficient lookups (binary search), then throw away d.
reply