|
Quote: Here’s the problem: people interpret SemVer [numbering releases using the a.b.c convention] as permission to make breaking changes. What's a breaking change? If, for example, it modifies a function's arguments but is well documented so that users can easily change their code, big deal. The other end of the spectrum is something that forces non-trivial rework in some applications.
Quote: If you need to make a breaking change to your API, it means you screwed up. Bzzt. At the trivial end, it means that the surface-to-volume ratio is being properly managed. No one can predict all future requirements. At the nasty end, it means consulting with users and giving them advance notice if a framework requires significant evolution in some area to better support new requirements.
This article strikes me as 🐘ing entitled whining. People can stay on the current release if they don't want to put in the effort to evolve their software. Either that, or they chose the wrong vendor.
|
|
|
|
|
On the other hand, if I use a library A that depends on library B and library B makes a breaking change, I need to wait for library A to update its code before I can update to library B.
|
|
|
|
|
Even worse if you depend on libraries A and B, which both depend on C, but each depends on a different and incompatible version of C.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Now known as Microsoft Edge: AI Browser the change of name does not appear to have been accompanied by any significant changes. "It's got electrolytes!"
|
|
|
|
|
The "Everything that's hip and cool browser"
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Kent Sharkey wrote: "It's got electrolytes!" What about dihydrogen oxynate, or whatever that stuff is called? That goes well in a web browser, right?
|
|
|
|
|
Almost 11 million internet-exposed SSH servers are vulnerable to the Terrapin attack that threatens the integrity of some SSH connections. Oh, SSH...!
|
|
|
|
|
Oracle recently announced that the MySQL database server now supports JavaScript functions and procedures. Who needs Structured Query Language when you could use a Silly Query Language?
|
|
|
|
|
These malicious packages - deploying cyberespionage backdoors and targeting Windows and Linux systems - were found circulating via the PyPI repository. Security experts expect the problem to continue. Was it spelled P-Y-T-H-O-N?
|
|
|
|
|
So this is how all those password manager (and other) security breaches occur? Someone is running Python code behind the firewalls, DMZ's moats graffitied walls?
|
|
|
|
|
A pernicious potpourri of Python packages in PyPI
I liked that, sounds like an Adam West Batman villain talking.
I’ve given up trying to be calm. However, I am open to feeling slightly less agitated.
I’m begging you for the benefit of everyone, don’t be STUPID.
|
|
|
|
|
While parked ahead of a pause in duties for the Mars solar conjunction, Curiosity put its Hazcams to another use. "That's my fun day. My I-don't-have-to-run day."
|
|
|
|
|
86-DOS would later be bought by Microsoft and take over the computing world. It's The Year of 86-DOS!
Gonna make a comeback!
|
|
|
|
|
Wow, it's like discovering a fossil of an ancient homosapiens ancestor.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Imagine how it feels for those of us who remember when we were running QDOS / 86-DOS!
(I believe that I still have the QDOS manual somewhere down in my basement)
Religious freedom is the freedom to say that two plus two make five.
|
|
|
|
|
trønderen wrote: (I believe that I still have the QDOS manual somewhere down in my basement) Lucky you. I would very much like to have back my original DOS 5.0 manuals.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Valve is finally leaving Windows 7, 8, and 8.1 behind. OK, *that* might get some people to upgrade
|
|
|
|
|
A deep dive into the text mode editors we had and how they compare to today's "And we liked it!"
|
|
|
|
|
I guess anyone who has been in the software world for 30-40 years could produce a list of functions or capabilities, asking "Why did we leave that behind? And this? And this? I would love to have it in today's tools!
To give one example (sticking to IDEs): The very first debugger I used could single step not line by line but call by call. That was really useful to circle in the problem a lot faster than going line by line. I have never seen this in other debuggers. (Most likely some of you will now say 'Oh yes! Xyzzy-debugger has it!' I am referring to debuggers I have seen). The debug information contains all that is needed for offering this, so it could be provided.
Another nice feature of this debugger was that you could tell it to do line stepping on every line from 300 to 1200, and then drop line stepping from line 400 to 800, but rather do call level stepping in that range (or maybe no stepping). I am missing that as well.
Do you have other "long lost friends"?
Religious freedom is the freedom to say that two plus two make five.
|
|
|
|
|
trønderen wrote: Do you have other "long lost friends"? Yes! I miss FoxPro 2.0 for DOS. That debugger had line by line source stepping. It was the first time I had ever seen such a thing!
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Wow, that brings back fond memories. Things were so much simpler then, and yet, oddly, just as difficult!
|
|
|
|
|
I follow you!
Then, I have a strong feeling that a significant fraction of the complexity is self-inflicted. We are much to eager to show our cleverness in using fancy, complex mechanisms of our chosen programming language, modelling methodology or whatever. It seems as if we consider even the very simplest problem as requiring multiple classes, abstract and virtual, multiple inheritance, and subclasses in several levels. The rules of the pattern chosen adds it share of red tape. We make identifiers long and complicated to follow coding standards requirements, which also requires splitting the solution, which in informal pseudocode fits well within a single sheet, in twelve different files, each with their file headers and copyleft statements, namespace declarations and whathaveyou.
It is possible even today to write simple, straightforward code to solve simple (sub)problems. But you will probably make your co-programmer scream out in horror over all the deviations you have made from commonly accepted coding style and modelling methods.
When I write small programs for my private tasks, I still program in a simple style, which is quite different from how I program when the code will be seen / used by others.
Religious freedom is the freedom to say that two plus two make five.
|
|
|
|
|
There are countless articles about Agile. And the use of agile. The usability. The statistics. Until the next last one
|
|
|
|
|
The men who designed and crafted the first practical integrated circuit (IC), the type used in the computers to guide the Apollo spacecraft that made the first U.S. moon landing, were also the unintentional fathers of Silicon Valley. Just a smidge more powerful on Artemis
Love the photo in the article of a chip on the Apollo - "consisting of six transistors and eight resistors"
|
|
|
|
|
While the computers may be more powerful, Artemis is actually a step backwards from Apollo. The Saturn 5 was able to actually land men on the moon and return them. Artemis can't do this.
|
|
|
|