Bill Meyer

The key phrase being "valid compiling." And even that is not without limit. In a large application, when the Unit Dependency Cycles (UDCs) reach some threshold, the build time begins to increase exponentially. The exponent has been, in my experience, relatively small, so things do not grind immediately to a halt. But even a 3% degradation can't be sustained -- eventually, a build takes minutes to complete. In Delphi, that is pretty shocking. The problem is in how to reduce the UDCs, since they derive in large measure from poor design. Reducing them is a matter of making real design repairs, and that is expensive. But if you think there is a brute force "solution" I would love to hear of it. I can readily imagine ways of obtaining more breathing space: faster CPUs, more compiler threads, more memory available for compilation. But all you get from those is a limited ability to delay the necessary repairs.

Whether it improves compile time depends on just how many Unit Dependency Cycles you have. In a large legacy project, it can be a very large number, and reducing that number can greatly help with build time. Also not, "build" not "compile".

Yes, and a euphemism for fragile.

Cooperative was a great euphemism. There was no cooperation unless the developer got everything right.

May not be possible in a corporate environment.

Like a few THz? 😉

I think you have the relationship reversed, and meant to write that CodeInsight is only operative with LSP. As reported above, the problem is present in RAD Studio 11. No doubt that is why the case cited is marked "unresolved."

Microsoft: Word, Excel, PowerPoint, etc.

When your products mostly have single word generic names, I suppose you can make use of installers which present no meaningful information.

And for me, for data entry. No email arrives, however. UPDATE: Second attempt succeeded.

Since variants 0 and 3 are reserved, then there really are only two variants. But there are still variabilities in Version. I'd wait to see the spec from IBM, and start be determining whether you can use the existing GUID support with some minor string manipulation afterward. Too many unknowns if you haven't seen their spec.

Have you looked yet at RFC-4122? https://datatracker.ietf.org/doc/html/rfc4122 The document you cite only hits a few points. One issue it raises is that of case, but in the small quote from the RFC, it does point out that although the UUID is output as lower case, it is case insensitive at input. Awkward wording, but I would take that to mean that the UUID is to be read in a case-insensitive fashion, whereas "input" could be interpreted to mean "from the keyboard" or other device. The RFC offers an example of a properly formatted UUID: urn:uuid:f81d4fae-7dec-11d0-a765-00a0c91e6bf6 It seems obvious that the issue of case should never be allowed to interfere, given that the alpha characters relate only to hex values. As to your question about a library, I don't know whether one exists. But given that Delphi inserts a GUID for you in this form: ['{480BE84A-F629-4A7C-BC94-4FC06A2E03BA}'] Then by stripping off the leading and trailing braces and brackets and converting to lower case, you would have an RFC-4122 conformant UUID which is of the form specific to MS. And for what it may be worth, I think the author made a poor choice enumerating the four types starting at 1, when the RFC lends itself more naturally to a zero-based reference. In the end, I suppose I should respond to your question with "what do you mean by that?" We could argue the semantics of descriptions, but in the end, the GUID conforms - per the RFC - to the MS variant rules, and the UUID subsumes the GUID.

My dislike for braces dates from a time when I frequently used printouts, and back in the day, it was all too easy for some of them to either disappear or be sufficiently unclear as to be missed. I am also not keen on some of the expression operators, but I seem to adapt more easily to those than to the cursed braces.

As many of us are likely to recall from the old forums, a key perspective here is that "developers don't understand marketing." In order to consider making changes, one must first be open to the possibility that the current state of affairs could be improved.

And yet, it is considered a member of the C family. And probably for no better reason than the use of those braces. And the braces are probably my least favorite aspect of that language.