Hmm... my experiences differ from yours. I find LibreOffice's styles very useful and predictable. I have 100+ of my business templates in Writer
1. F11 to pop open the Styles and Formatting sidebar (undocked is my preference)
2. Use Ctrl+M to clear direct formatting.
3. Keep a master template of styles to re-import into the others to ensure uniformity. (Make sure to check Overwrite or you're wasting your time.)
4. Don't confuse paragraph styles with character styles. (Use Page and Frame styles for extra credit.)
Of course, NO word processor is going to do layout like a true graphics workhorse like Illustrator, InDesign, Scribus, or Inkscape. If you are looking for 1200 dpi accuracy, LO is the wrong software. But if you're looking for general office functionality, I don't know anything better, free or not. (And my first word processor was Claris MacWrite II about 1989, with long stints of WordPerfect and Word up until 2013 when I went pure OO.o/LO.)
Except that's the way professional CAD packages do it, for example.
You keep mentioning CAD software, but there is no similarity between CAD formats and typical raster formats. JPG, PNG, TIF, and GIF are industry standards that GIMP should be able to open and save. CAD formats have vectors and a lot of other information that is particular to each software's implementation since there aren't open, widely-used vector file formats.
There are AutoCAD-compatible clones that open and close that proprietary format without question. LibreOffice also opens and closes non-native formats without complaint if you turn off a checkbox preference. If you stretch the analogy enough, we can take this illustration to even further ridiculous extremes. But the fact remains, GIMP is raster software that can be used to edit lots of standard format raster files that shouldn't require exports for files opened.
If you honestly believe Export is the correct terminology and UX, then you should also be advocating GIMP's "Open" menu change to "Import." That would correctly match what you are claiming on the save side. Open would then be reserved for XCF files but everything else should have to be imported. GIMP should treat the whole business like RAW and refuse to ever save any changes over the originally opened file unless it is an XCF.
At least that would be consistent. Insane, but at least consistent.
The current GIMP behavior appears to be structured to sell the software's (developer's) capabilities over users that are too stupid to think for themselves. Unfortunately, everybody sees this except for the developers.
Engineers are responsible for the stuff they build.
Right on target here. Ancient laws held architects/engineers to the standard of an eye-for-an-eye or death for designs that failed to the result of equivalent injury or death. I'd like to see a similar system (another post this thread) that licenses software engineers and architects.
Most of the public do not realize that local authorities hold architects primarily responsible for criminal and civil damages instead of themselves. I'm not allowed to break the law, but in cases where it is vague or interpretive, my seal stands as the primary guaranty of safety over the authority having jurisdiction. "It's your seal!" they tell me.
Licensure is the public connection to responsibility.
Not sure what you mean. Architects don't design engineered systems and engineers don't practice architecture. Doing so is outside of either's scope of licensure and expertise. The occasional architect or engineer attempting to practice the other's will pay fines and can loose their license. I can't tell you how many projects I've cleaned up on behalf of Owners that were started by unqualified and unlicensed "designers and engineers."
Design is the comprehensive practice of conceptualization, strategy, and coordination. A myriad number of complexities get sorted by architects that may or may not involve math, as you say. Component and material systems and connections, building functional accommodations, adjacency efficiency, occupancy loads, exiting, fire resistance, accessibility, material finishes, acoustic performance, health department demands, lifecycle costs, envelope design, energy performance strategies... just this week for me.
Specifically to the article topic, it's the same for software. There are both design/architecture and engineering components. Design tools focus less on math because they are more strategic. The error is saying one is more important than the other.
I think your industry would do well to create a vetting system that established core competencies for professional software engineers. Anyone so established via education, training, and testing would be distinguished from the large mass of us who hack evenings and weekends and who really shouldn't be entrusted with protecting the health, safety, and welfare of the general public.
Licensure is authority established via the government, usually each individual state. There can be reciprocity between them, but the it certainly isn't a job or social title. Doctors, lawyers, nurses, accountants, and architects are all professional in the powers vested by the state.
Lest ye protest, that's the way it has been for ages with building engineers who design plumbing, mechanical, electrical, and civil realms in our physical environment. Why should the software components get an exception? Why wouldn't those of you who are actually competent want to establish a base level of expertise that helps sort this for the rest of us a bit?
My own industry of architecture is the same. You can't call yourself an architect unless you meet the requirements of the licensing board, about 10 years education, training, and seven tests (recently changed from nine). Of course, you software guys have been trying to abscond with the term for decades, even as a verb, but it is a self-proclaimed title and means nothing because there isn't any vetting. I'd be fine sharing it if it actually implied both the comprehensive design nature of what architecture means and professional licensure, but it will always produce an eye roll from those few of us who have been through the decade of sleepless nights and rigorous vetting to be a real one.
"How to ask a question the smart way" is 23 pages long and starts with the presumption that the questioner do most of the work in solving the question prior to asking.
As long as we spend more effort on Slashdot explaining to a disabled, non-developer that he is wrong than it would take to fix the bug, it is will NOT be the year of Linux.
IAAA. Forget robots, Warner Sobek's R129 house is genius.
Check it out in the PBS e^2 series, part 6/6, at 20:07.
Back up to 18:55 to see the beginning of Sobek's ideas.
Watch the whole series if you have the time.
In our architectural practice, we use the term "additional services" to quantify scope creep, basically anything beyond the scope defined in our proposal/contract.
This reinforces the need for making that initial statement of expectations clear AND the implications for any deviation thereafter.
What this country needs is a good five cent microcomputer.