Actually, I like Windows, like it a LOT! Hold on ..., some strange circumstances got me to get Windows 10 Home Edition on a laptop computer, and I don't like W

Actually, I like Windows, like it a LOT! Hold on ..., some strange circumstances got me to get Windows 10 Home Edition on a laptop computer, and I don't like Windows 10. I don't like the undocumented new behavior or the many updates and changes. Still, I like Windows, but just why and especially how, I will keep a secret so that the how of my secret doesn't get changed! Office? I LIKE Office! I like Excel a little (generally find it easier just to write code in a procedural language than to use Excel) and especially like Outlook. But the version of Office I like is from 2003. Due to a change in SMTP email standards, I will have to upgrade to 2007 Office (from eBay) or some such -- that's on my TODO list. Except for that SMTP change, I'd prefer to stay with Office 2003. I know how to use Office 2003, and I don't want to invest the time, and frustration, especially the frustration, to learn about changes, any changes. To me Office is NOT a very good tool, NOT at all central to my work in computing. I do NOT believe that Microsoft has found any worthwhile changes to Office (except responding to the SMTP changes). Net, I do NOT believe that there is anything about new versions of Office that are worth my time or frustration, especially the frustration, to learn. If I want to do something Office 2003 can't do, then I'm perfectly comfortable writing some procedural code, usually for work there is no hope of ever getting done in anything like Office. E.g., I had a 0-1 integer linear programming problem with 600,000 variables and 40,000 constraints. I found a feasible solution guaranteed to be within 0.025% of optimality, no hope of ever doing that on anything like Office. I see a larger theme: People who have a popular product want it to keep up so work hard to make changes, change or delete old features and add new features. This theme applies quite broadly, e.g., to cars, Windows, Office, women's fashions, suburban single family home architecture, ..., and Web sites. Presumably those people hope, expect, or whatever, that the new version will be better; in fact, too often the new version is worse. Yes, when WiFi, USB, etc. come along, we can need corresponding new functionality -- no question for such cases. But such changes come along only slowly, and changes to user interfaces can happen nearly everyday at nearly every Web site in the world -- biggie issues and questions there. For cars, e.g., I have an old GMC Yukon, in good condition, and it has some dozens, maybe hundreds, of features that seem to need a computer to do the logic. So, the Yukon keeps trying to get my attention with various lights and sounds, new to cars. For the new stuff, I don't want it. Apparently the GMC designers rushed to conclude that with computers there was a lot they could do to manipulate sensor data and generate lights and alarms. COULD they do that? Yes. But SHOULD they do that? Nearly always, NO! Here's a broad example of this theme that has a lot to do with 3+ billion people, literally: Way back there were many, millions, of different paper forms with boxes, places to make check marks, etc. Then there was some user interface work, via both hardware and software, by IBM for their 3270 series of data entry terminals, work that borrowed a lot from such paper forms. Since people understood paper forms, both the programmers and the users found 3270 relatively intuitive and easy to use. Then there were dumb terminals, i.e., asynchronous ASCII, and there were a lot of programs to drive such terminals and that borrowed a lot from 3270. Then there was HTML, and I'm guessing that it borrowed a lot from all of paper forms, 3270, etc. As we know, with HTML there were controls -- single line text boxes, multi-line text boxes, radio buttons, check boxes, push buttons, and links. As I recall, that is all, ALL there was. Want to build a Web page? Okay, are stuck with just those few, simple controls. Now we come to two points: (1) Those controls were so close to, literally, paper forms that they were automatically relatively easy to use. (2) Literally BILLIONS of people, necessarily nearly everyone in the world making use of the Internet, 3+ BILLION people, understood how to use a Web site with Web pages using just those HTML controls. In important respects, the HTML controls, borrowing so much from what worked so well in the past, were VERY well designed. Then with the discipline of just those few, simple HTML controls, every Web page started with a HUGE advantage in user interface ease of use. It was still possible to build an awful user interface, but, just from the discipline, it was also easier, relatively easy, to build a good one! Then we got icons: Since icons essentially need a graphics terminal, couldn't have icons with 3270 or the dumb terminals. So, when we got graphics terminals -- terrific for sketches, pictures, movies, games, PDF files, Zoom, etc. -- we also got icons. BUT: When a new icon is designed, only God and the designer know what it means. Given an icon, can't spell it, type it, pronounce it, easily communicate it, describe its function just from its appearance, sort it with other icons, search for it, look it up in a dictionary, etc. Still, we got lots of icons! We STILL have lots of icons. Civilization HAD icons and then about 2000 years ago made a HUGE step up, the Roman alphabet. Now it dominates communications around the world; a lot of people even in Asia and Russia understand it, and otherwise essentially everyone who can read understands it. Then along came JavaScript (JS): With JS it was possible to build Web pages that ignored the discipline of HTML, have the pages very active, that jump around, have lots of secret functionality, etc. It was possible to have an ocean of new functionality. Alas as we would expect from the theme and history, a LOT of the new functionality was very unwelcome and ruined a LOT of user interfaces. In particular, now each Web page done heavily with JavaScript likely is unique, has to be learned on its own. In simple terms, with just HTML, there were many millions of Web pages relatively easy for billions of people to use right away, and with JavaScript suddenly there were millions of Web pages all different, often poorly designed, and that had to be learned one at a time. E.g., now a single line text box can be invisible until something else happens at which time suddenly the box becomes visible -- the user has to learn this undocumented, unexpected, unprecedented, obscure, bizarre behavior just by experience, frustrating, error-prone experience, uniquely for nearly each such Web page. Nearly always, similarly for icons -- learn by experience. Silicon Valley must have mountains of dedicated, dead ex-Google Web page designers and JavaScript programmers who died miserable deaths trying with all their strength to make massive changes to the first Web page at Google! Instead, Google has kept that page, at least as a user first sees it, both OLD and SIMPLE. I credit 80+% of the Bezos fortune to some discipline, from somewhere at Amazon, for some relatively good and SIMPLE user interface design -- people around the world shop, place orders, and spend their money right away! I would rate the Amazon user interface as in the top 1%, by wide margins, of the shopping Web sites in the world -- we're talking billions of users and about a $trillion in business worth from good, relatively simple Web page design. Net, in short, just because some new technology makes some new feature possible does not alone make that feature good.