Smug As A Bug

[yogi]

Today was a test of how well I learned things. The final stage in this project is to put the ASUS tower software back together as it was before UEFI became a concern of mine. If you have been following the saga from the start, you will recall that I began this adventure with a system configured for MBR booting on an ASUS motherboard that was essentially designed for gaming. That is to say, it's not a dedicated Linux box.

Seeing the end of the line for Windows 7 made me take an interest in a possible alternative. Thus I set out to learn about UEFI which is the next step for Windows but also the future of Linux and any other desktop operating system. I honestly cannot say that I have UEFI booting mastered, but I know enough now to get myself out of most trouble situations. With that kind of confidence I deleted all the MBR OS's, except for Windows. I have no inclination to reinstall and rebuild Windows 7 after I just spent ten years customizing it to my needs. Thus, I migrated the existing Windows OS installed on an MBR disk over to a blank GPT disk. After some fears and trepidation, the migration was a success. The tower boots UEFI and retained the two partition schema I originally installed. I also use an encrypted internal data drive which was a no-brainer to move to it's new location in the GPT world. The only remaining task was to install Linux and boot it UEFI alongside Windows.

You know how many articles are written describing how to do that kind of installation; I'm sure I posted half of them in these forums. LOL So, today would be the payoff. I chose Ubuntu 18.04 as the first Linux OS because it's the LTS version. The later versions actually work better in some respects, but I wanted to get back to were I was before all this UEFI business. The installation went well and no glitches that I noticed came out of the installation. The big question with Ubuntu always has been where it decides to install Grub. In the old system it decided to overwrite the Windows bootloader and was one of the reasons I decided to learn more about UEFI. It turns out Ubuntu's installer is flawed. In any case, this installation would be the test not only of the Ubuntu installer but also of my skill in fixing any problems it might cause.

When the installation completed I did the usual reboot and ... well ... no Linux. Windows came up in it's place.

It was good to see Windows, but there was no menu in BIOS that allowed Grub to appear. This is exactly opposite of what I have been seeing in the past. As it happens, the USB Linux on a stick version booted without incident. Using that stick version I did some investigating. The ESP boot partition was empty. This was scary because that's what happens when Ubuntu overwrites Windows. However, it did not overwrite anything. It simply did not install. The ESP partition could not be recognized by gParted either. After some consideration and review of past experiences, I made the assumption that the Ubuntu live image did not boot in the UEFI mode as expected. It booted MBR and, of course, the tower no longer recognizes MBR. I rebooted the Linux on a stick installer and discovered there were two choices among the several my BIOS was offering. One of those choices had UEFI in it's descriptor while the other did not. I chose the UEFI version and the installation was only slightly different. The good news is that Grub was written to the correct ESP partition, i.e., not the Winows ESP but the Linux ESP - both on two different disks. That is exactly the schema I had with MBR.

Apparently when I installed Ubuntu the first time around I did it in the old BIOS mode and not the new UEFI mode. I know very well UEFI cannot be installed from MBR, but I failed to recognize the minor difference in the installation program that gives it away. Ubuntu does not make it obvious which mode you are in.

So now I am a happy camper. My tower is working as expected, but in the UEFI boot mode. Ubuntu STILL has the problem of writing Grub to the wrong ESP partition, but only under certain circumstances. It wrote to the correct one today. Unfortunately, Linux Mint and Linux Peppermint STILL identify themselves as UBUNTU. That means they cannot be installed on the same machines unless you want to go to EFI/GRUB Hell. Since I like both of those distros, the only way I will be using them is when they are installed in a virtual machine or as a stand alone Linux on a stick.

Life is good again.

[Kellemora]

Well, CONGRATULATIONS Yogi!
You now know more than those who create the programs in getting them to work properly.
My HAT is Off to you my friend!

[yogi]

*takes a bow*

Thank you for your accolades. I did indeed accomplish a long standing goal. However, I did nothing that is beyond the normal capabilities of Windows, Linux, or UEFI. The standards are pretty clear about what they can do. The so called "experts" are proficient in what they do, but seldom do they know anything outside of their little boxes. This leaves a lot of questions unanswered, or worse, a lot of wrong information in what is supposed to be tech support. I guess I should not expect anything else given that all those support forums are user driven and free of charge. In order to get some expert help one must pay. There are exceptions. The two guys who you pointed me to in the Linux Mint forums (regarding Linux on a stick) were obvious experts. They were extremely patient with me and very helpful. My understanding of UEFI boot could not have been accomplished without them - too many conflicting stories.

I've run out of projects for my Command and Control Center. The end of Windows 7 is the end of an era. I have only one aspiration for the future. I want to build another computer with super powers. It will be costly and state of the art. I don't know what I'd do with it, which is the main reason I've not built it already. Then again, who knows what super powers will be available the next go round.? Maybe I need another hobby ... I've been watching a lot of quilting programs lately.

[Kellemora]

You can laugh if you want, but I used to do a considerable amount of quilting over a five year period.
I was partway done with the Tree of Life quilt when we had to move, and although I had room, I never got back to it, too many other things on the fire to make money now that I had a place where I could.

Maybe you could build a Quantum Computer with AI and let it build all the rest for you, hi hi.

My new neighbor across the street who runs Windows might be doing the same thing you are doing, and getting pre-release updates.
Although I have no idea what he's talking about with his Redstone, 19H2, and 20H1, he's currently testing a Windows10 version for ARM64, had been for some time now too or so he claims.
He decided not to do whatever it is my other neighbor is doing, said it is really above his head, and too costly. But he loves having access to the Fiber Optic cable through him.
He must have money to burn, he says he is connecting to his neighbors fiber using a pair of 802.11ac wifi routers, but has a pair of 802.11ad wifi routers on order. Over 5 gigs with the first, supposed to be over 7 with the ones he has on order.
I have no idea where he and his wife work, but I only see them home at night and on weekends. So why he needs all of that is beyond me.

[yogi]

Your neighbor is in the same beta testing program that I am in. In my case I switched to it very early in the Windows 10 release cycle and never left. Redstone is the code name for the stable release of Windows 10. The 19H2 software is for the current feature release while the 20H1 is the jump-ahead version for the next generation. Since the last major feature release in October of 2019, all us "Insiders" have been on the 20H1 flight path. If Microsoft holds to it's tradition, the stable release would be available this coming spring. Until that time us pre-release testers get updates once or twice a week. These are not the ordinary bug fixes and security updates that everybody else is getting. They look and feel like the real thing and only on rare occasions have I experienced minor problems.

Microsoft has been loath to support ARM processors. These are a special class of reduced instruction set processors that you find in the smart devices you might have laying around your house. I'm not sure, but I think they also are in some smart phones. In any case, not many people have an ARM processor in their computer and there was no interest on Micorsoft's part to capture that market until recently. I guess it's like their invasion of Linux. They want to be everywhere.


The public television network is all I watch on television, if and when I watch television. I'll generally do that right after I wake up in the morning and before I make myself some breakfast. They used to have cooking shows and painting shows on at that time of the day. I loved the cooking shows and was only slightly interested in the artists. A few weeks ago they changed their programming. Now the morning shows are dominated by sewing and quilting and some other drivel that must have been on sale in the broadcast show stores. About the only "interesting" show on PBS now is the yoga one. It's pretty fascinating stuff to watch, all those ladies in tight fitting clothes and all. I don't see how they can call that exercise, but apparently yoga is very popular among certain groups of people.

[Kellemora]

I guess he probably is!
Many eons ago, when we had the WANG VS, Wang would give us a list of things to run on the computer about once a week and then save the report for them to examine. This was over and above all the things they were doing on it themselves.
It was mainly testing certain areas of the computer by doing I guess complex jobs. And as usual, I had no idea what the coded info I was typing in meant or what it did for that matter. I do know a couple of those seemingly simple things bogs down the computer for about 5 or 10 minutes, but usually only if they had things running using up a lot of the resources.

I don't know what ARM is, but assume it must be some type of computer system for Schmartz-Fonz.
Actually, Windows adding some Linux functionality to their computers, appears to a win win on both sides of the fence.
Linux users may end up getting some proprietary software for a price of course, but it will be locked so you need the license. And there may be a file created just so proprietary software has an area to store certain data, like timed use, expiration dates, and the like, in an encrypted folder owned by Windows.
It seems the reason no one has ported their proprietary software over to Linux was they had no way to protect it from users changing the usage time allowance or free trial timers. But if they add this encrypted file folder as a place to put such info, we may see more proprietary programs made available to Linux users. But like everything else, Micro$oft wants full control of this folder, and developers have to go through them to get the codes set in place.

I rarely watch TV. I normally sit and work a jigsaw puzzle in the TV room while a show is on.
Have to spend some time with the frau each day, and after dinner works well.
Then I boogie back up to my office until bedtime.

[yogi]

The ARM processors require special code and must be compiled as such. My guess is that a lot of developers choose x86 or ARM but not both. It would be akin to writing software for Windows or for Linux. You may get something to function the same on both machines, but it's not the same code. That's why you don't see too many Windows classics on mobile devices. It would be like having your development team make stuff for Apple. LOL

Windows tried their luck at breaking into the mobile market and failed miserably; more than once even. So now, I presume, they are just accepting the fact that they can't compete with all those ARM-ed devices and will simply join the club. There is no need to segregate code or file structures. If you want to run MS Excel on your Apple iPad, you simply will have to buy it, or rent it. The trick is getting Microsoft to write the code in the first place, which is what they are becoming interested in doing these days. You know, if Libre Office can run one Windows, why should MS Office not run on Linux?

[Kellemora]

Actually it can. I had a ton of spreadsheets that would not load properly in Open Office Calc, so I loaded msOffice through WINE and it loaded up OK and opened the files. From there I could save them in another format which Open Office Calc could understand better. That was a long time ago now, so I don't remember what options msOffice offered at the time anymore.
I also have hundreds of WRI files, but they convert to RTF easily enough.

By my way of thinking, I don't see why msOffice couldn't be ported over to Linux or ARM for that matter.
Programs don't speak directly to the CPU anyhow.
They speak first to the Operating System and then the OS speaks to a Compiler or Assembler, which in turn speaks to the CPU in binary.
Seems like a compiler could be made with input ports to accept Linux or Mac or MS program coding, and the output could be to the type of CPU whether it be X86, ARM, or whatever.
Also by that way of thinking, it seems possible also to have a compiler for each type of input program, so you could run Windows, Mac, or LInux software on any computer.
Early software that did have contact with the CPU might not work so well, like using Peeks, Pokes, and Calls in Basic Programming for a 6502 Microprocessor. But the code could be altered to work on 8080a processors.
However, doing things that was sorta stopped many years ago as computers got more complex and could do more things from the upper level humanly readable code.

[yogi]

The idea is to get central processing units to understand human readable code. If that's all there was to it, then your thoughts about a universal compiler would be easy to implement. But, as it happens, CPU's are not all equal. There is nothing that can translate all programming languages into binary that all CPU's can understand. If quantum computers every get to the mobile device level, then technology at that point may be advanced enough to allow for what you suggest. There are just too many variables in today's world to cover all bases with a universal programming language. Besides, if all computers were equal, there would be no glitz factor and everybody would have to charge the same price for their machines. That will never do in a capitalistic environment.

[Kellemora]

I didn't mean ONE intermediary program to speak to the CPU. I realized there would have to be one for each type of CPU, and the OS would have to be written to account for that. However, there could be more than one intermediary program associated with each OS so it would select which intermediary program to use on which computer. That way, once you had the OS, with all the intermediary programs built in, you could install it on anything.

I do see more and more proprietary programs being offered for Linux machines, but have no idea how they are protecting themselves yet. I do know a few are by subscription, so the program probably checks to see if the subscription is current before starting, which means it only works if you are on-line.

As an aside: I used my old XP program and installed CARDFILE.exe on it, to possibly resave them as text files.
It let's you do that, but the output does not look any different than reading the .crd file with notepad, very convoluted.
I also tried the newer version of cardfile, forgot the name of it already, and it did the same thing.
Also, I really no longer need the data from most of those cardfiles I saved. Most of them were addresses and phone numbers for business and individuals, and nearly everyone dead or the businesses closed or changed hands and the person I would normally speak with is no longer there.
It was still interesting to look back at all the companies I dealt with almost on a daily basis.
I also had crd files for nearly every organization I took care of the paperwork for.

I found LibreCAD can open the files I made with AutoCad and AutoSketch, and of course the ones made with QCAD.
I think LibreCAD has replaced QCAD in most Linux Distro's now.

[yogi]

I think I understand the beauty of having one machine that will run any program ever invented. I'm fairly certain that can be done with virtual machines, but it depends on the host. Wine on Linux is a good example of why it isn't practical, albeit Wine is an abstraction layer and not a virtual machine. You can only go so far mixing and matching dissimilar processor functions. I also don't see why you would want to run several different versions of, say, a word processor. You said yourself that you don't like the way Microsoft does things. Even if you could run MS Word on Linux, would you want to? LOL It might be a great engineering project, but I don't see such a computer being used by many people.

At one time I had the notion that I would update my contacts, the modern day version of cardfiles, and put all the information into a database. I'm fairly certain it wouldn't take much to get the date from the cards into the LibreOffice Base data store. Then what? That's where I got stuck in spite of the experience I had using Oracle's database software. The idea of putting it all into an SQL database makes it timeless. I doubt that SQL is going to become deprecated in my lifetime, although I know there are groups trying to do exactly that.

[Kellemora]

You've seen me mention how much I loved the original msWRITE program.
One of the reasons I fell in love with this program was because of its global increase or decrease feature.
Such a feature is not found any other word processor that I know of.
Whenever I had to create a flyer I always started with WRITE first, then moved it to DOC later if need be.
The WRITE program is simple and still works in WINE today on Linux, but I rarely do flyers anymore, and have found workaround in LibreOfficeWriter by using classes.

As far a Databases go, I used to do a lot of things in a Database, but it later years simply used a Spreadsheet.
I realize a Database has some great features not found in a Spreadsheet, but I rarely had to set up a new Database after creating one. So what happens is I forget the nuances of creating a new Database, which MUST USE and ID column.
Now, although I could set the Database to add new ID numbers sequentially so I could forget about that added step.
If I wanted to add a new field for some reason, I forgot how by the time I needed to do it.
For some reason, I never could get it through my head to create a Table first, hi hi
On newer programs, you can simply hit EDIT and it will open up the Table creation in edit mode.
But back when I was using them, you had to select Table or Field. My other worry was the Encryption selection, which I never did fully understand. If I set it one way in one table, then set it another way in another table, I could not merge the two tables. This is one reason I started using Spreadsheets instead.
But it has been years since I've looked at or used a Database. Maybe I should look into them again.
That being said, I do use FreeMind which is a Mind Mapping program I've become very comfortable with.
FreeMind will let me include links to the actual documents, so if I can't remember where a particular document is filed, I can use FreeMind where I have those types of documents sorted together and just click on the link.
Works great until you move files around, hi hi.
I do remember I saved my Database as GUI, so don't know if it actually used SQL or not.

[yogi]

All my experience with databases is in SQL. I tried mastering that which is included in Microsoft Office, but had no luck. Setting it up was not a problem, but doing queries was absolutely insane. I was barely able to use the Oracle database we maintained at Motorola, but that one was logical. All that being said, the fact that you can vary the queries to suit your particular needs of the moment is the big attraction. I suppose there are methods other than SQL when dealing with a database, but SQL has an enormous amount of flexibility. I suppose that's why it's so easy to hack. There are so many ways to do it.

[Kellemora]

There are a couple types of databases I've used in the past.
One was directly associated with msWORD to fill out the names and addresses on form letters and the print the envelope.
In this one you just had to fill out the fields and it would pull them in.
Trouble is, if I recall, I couldn't select which ones, it was all or none, unless I manually removed them.
The second one was a big normal database with all the names and addresses of everyone who belonged to the church.
This one you could go in and set which ones received something like a letter or newsletter, etc.
There were also ways to keep them sorted by exactly what type of letter they would get, or what purpose they were being mailed something.
Naturally there were fields for birthdays, anniversaries, etc. and it was easy to only query for birthdays that week or month to send out the cards.
I suppose this could have been a SQL database, but it looked just like GUI database I used a few times myself.
The whole thing about doing queries is that you have to set up the initial table with those search criteria fields.
Which means you also have to be able to have filters to sort certain fields, like birth dates.
You could set it up with a single field for each week, or each month, but it is easier just to have all birth dates in a single birth date field and use filter to pull out one for a certain week or month.

[yogi]

SQL, the language, is not as simple and straight forward as you describe. You were using some kind of front end that could have been using SQL to get the data you needed. You can actually make GUI web pages from SQL and query statistics from there. Doing any of this effectively requires knowing the structure of the database itself and then knowing the right SQL commands to get the information you want. I've milked our own database from time to time using a built in SQL query panel that writes the commands for me. But, when I learned how to do all this, I was creating databases and the SQL queries from scratch. It's been so long ago that I forgot almost all of it. The databases you see in the Office Suite software is good enough for most people, such as your church group. It's terribly under powered for anything like running a business, or even running a website like this one. I thought it would be a great way to keep in touch since I don't happen to have an Oracle database at my disposal. LOL Well, I do have mySQL and I suppose I could learn a lot if I tinkered with that. It's the preferred tool of hackers, but I'm not in the hacking mood this year.

[Kellemora]

I belong to a small website group that technically started around 1989.
The entire website was written by a college kid learning programming and he set it up as a hobby to keep active in certain types of programming. He kept it growing and adding to it, until he had a full roll-out of his completed website around 1997, but it didn't become active until 1999 when he had all of us from his previous website move over to the new one. He couldn't afford to keep two going and pay for his college debt after graduation.
Like me in a way, he did things his own way, and as it turned out, not the direction the crowd of programmers migrated to.
The way he designed his database worked for him, but was so unconventional by later standards, there is no way to move it over to a newer more modern type of system. The thing about the way he created his database was actually excellent and perpetually expandable. Trouble is, nobody but he understands what he did to make it work so well. And for this reason, nobody knows how to extract the information from it over to a modern database. He has had many try.
Like I said also, he started this as a hobby, and although he stuck it out for many years, he finally passed control of it over to one of the members who knew just enough php to keep it running, but not how to change any of the base programming that drives the whole thing.
Despite all the changes in computing over the years, the website still works great.
The current person running it has also hired programmers to look at it to find a way to migrate the database.
The problem they say is that it is not a database at all. It is more like stacks of folders addressed through a programming language such a C++, but it was something else he used there too.
The last time I talked about this with the current operator of the website, he said it is like a gigantic spiderweb.
When a new member joins, their screen name goes into one folder, their real data into like three other folders, their password divided into six or seven different folders each with a serial number but not a sequence number.
Once logged on, if this person wants to make a new post it goes in another set of folders. If they reply to existing post, that reply is in yet another set of folders. And all of those have to be read to display the original post and each comment under it, and in order. And nobody knows how he managed to get it to do that. But on the bright side, no hacker would be able to figure out what someones password is, because it is not a string located in a folder somewhere, hi hi.
php is used to make the html page you see when you log into the system, just like most larger websites.

[yogi]

That pseudo-database you describe sounds amazing. LOL It's a concept I've actually seen used while I worked at Motorola. One of the technicians who didn't know much about databases wrote a defect tracking program in BASIC that stored information about build failures in a similar fashion to what you describe for that web site. Each incident might have contained a dozen or so pieces of data with information about the defects being recorded, and each bit of information was in a folder of it's own and various subfolders. The net result was a huge amount of memory being used to save the data. In fact that amount of data exceeded the inode count of the operating system. The machine crashed one day and it took quite a while to figure out why. I didn't know there was an actual limit to inodes but apparently there is.

I've never tried to understand how a database is constructed, but it's got to be similar to a spreadsheet. I believe the MS Office spreadsheet program has a 64,000 line limit to the amount of data. I'm pretty sure the column count is similar. That's trivial for databases containing millions of records. How they can jam so much information into such a small space is beyond my ability to comprehend.

[Kellemora]

A database and a spreadsheet work almost the same way, except a database requires an ID number for each entry line.
The reason for this is you may have the same name used several times. Like John Smith with different details than another john Smith, and even another John Smith.
In a spreadsheet, the lead column must contain a different bit of data, but a database the lead column is an ID number.
Other than that, and the pre-selected filters, they are almost identical.
Spreadsheets are only like 16,000 wide but over 1 million lines long.
A Database is limited to 1 million rows also, UNLESS you do it in SQL then it is 64 Terrabytes, if your system can support that much, hi hi.

[yogi]

As far as I know there is no such thing as data tables to store information in a spreadsheet. You can indeed construct things like Pivot Tables but all the data remains in that container. Not so with an SQL database. There are tables (containers) for everything and related items are coordinated by index numbers. For the most part I never saw or used the index because it was automatically created by the database software. I could define what it would be and how to increment it, but the actual data and not its indexing is what we manipulated. I never had to construct a database from scratch but I did go to a class where they showed how to do it. A list of business contacts would be relatively easy, but keeping track of corporate information is more than a single person can handle.

When I used MS Office software they made the database look like a spreadsheet. That was a nice way to organize things but it was deceptive. That's not how the data is kept in storage. There are no rows and columns other than what is contained inside each table. Querying several tables would produce a report of line items that matched the criteria, but that as a construct of the report generator and not the way the data was actually stored.

I'm getting a headache thinking about it. I'm glad that I'm not doing that anymore.

[Kellemora]

I was a draftsman from 1966 to around 1978 when dad had his heart attack and I had to go take over for him in the family business. In 1980 when he returned to work, I started looking again in the field of drafting, only to find it had changed considerably due to the advent of computers and computer aided drafting. So I stayed at the flower shop until we closed, while at the same time putting together everything I needed to open my own hydroculture business.
This is also about the time I bought AutoSketch to get my feet wet so to speak on an affordable cad/cam program. There was no way I could afford to buy AutoCad yet. I was hired to help with a couple of historical home renovations which I did while the flower shop was still open. Earned enough money to buy an entry level AutoCad program. Turns out it was quite different than AutoSketch, but the basics were still the same. Oh, the reason I bought it was because most of the work I was doing on the historic homes was done on AutoCad and they could give me disks with the data on it.
Then when I started doing my own renovations, the architect I used also used AutoCad, and gave me the disks too, so I learned a lot just from seeing how they did things. Having AutoCad became a boon to me while I was doing home renovations, especially when I was changing the interior designs around and needed to get permits.
AutoCad had three major revisions and each one cost an additional 350 to 500 bucks to upgrade what I had.
I was away from it several years since most of my renovation work was simple. Also computers changed a lot too.
When I started doing some remodeling on our new house down here, I looked at AutoCad but couldn't afford it.
I was also doing a lot of my work on Linux and tried QCad, it was more like AutoSketch than AutoCad, but sufficed for my needs good enough. It is what I used for all of the remodeling to this house, which was extensive. Load bearing walls removed, etc. Lots of major work went into this house. Only the minor stuff is left to do, but after two heart attacks, well, that ended that!