Hacker News new | past | comments | ask | show | jobs | submit | graycat's comments login

I used to live in New York State and, back before the Internet got very powerful, got frequent offers from the NYT to subscribe.

I had long since concluded that the NYT was junk, resented the offers, and wrote a scathing denuciating letter to the Publisher ending with I don't want your work for free.

Actually, I was a little too hard on the NYT: On paper it can be used to wrap dead fish heads.

My short recommendation for the NYT and the rest of the mainstream media is, first, to get the level of quality of writing and information up to common standards for high school term papers.

Yes, I was in IBM, some AI in their Research Division, from 1985 to 1993 when IBM had a big crash.

Maybe an old idea I had about PCs has some insight: The idea is that for several years into the rise of the PC, the base, solid as concrete, fundamental, economic productivity reason for the PC was to kill off the typewriters, that is, word processing; as one guy put it, "capture the key strokes". The typewriters didn't "capture the keystrokes" and, thus, were a huge economic waste. Next in line was spreadsheets.

Now the biggie? Okay, replace TVs. We've already essentially replaced newspapers printed on paper; and PDF is replacing a lot of books printed on paper.

The future? See a problem that in terms of economic productivity needs solving, and get one or a few PCs and solve it. How? There is nearly no limit on the new problems to be solved or the new means of solution.

On several points I agree with the OP on the desirability of "universal computing". For more, I've long thought of most of the concerns in the OP and have not been and still am not much concerned: Why not?

=== No Smartphones

For all the threats of smartphones, I avoid them -- I don't have a smartphone. My phone is an old Bell touch tone desk set connected to essentially a land line.

I saw some threats of smartphones and didn't like the cost to buy, cost of usage, bad keyboard, small screen, and the general inability to write and run my software, old and new.

=== Digital Appliances

For threats of digital appliances, devices, from, say, Amazon, I don't have any. No way do I want some digital appliance listening to everything I say.

=== The Cloud

I don't trust the cloud. I make no direct use of the cloud. So the cloud is not a threat or cost to me. And I don't have to do mud wrestling with their poor or missing documentation on how their services work. What they are offering, no thanks. Not even for free. And whatever they say about reliability, security, or functionality, I don't believe it.

=== Encryption

For threats of encryption, if the situation gets serious, say, for my email, then I will make use of PGP (pretty good privacy). So, I will have encryption under my control that Apple, etc. can't do anything about. And I won't have to worry about back doors.

There was a reason PGP was open source -- to keep a big power, government, company, from getting control over encryption, putting in back doors, etc. I like the idea of being able to control my own encryption.

=== Text and Console Windows

To me, the main data I work with is just text, the standard ASCII character set. And my main user interface is text in console windows. A big reason is that it is easy to automate the use of text in console windows.

So, in particular, I make minimal use of the Microsoft user interface idea of on-screen direct manipulation graphical user interfaces (GUI). I never liked the idea of a GUI -- insults me as a user; is an interface I essentially can't program; gives output tough to process further.

In one sense, important to me, a GUI is nearly always a big step backwards; it has me do something once; but to do it 200 times I have to do it 200 times as I do it once. Instead, I want to automate doing that 200 times. E.g., I had a list of about 300 URLs and wanted to download them all. So I used my text editor KEDIT to develop a REXX script to call the program CURL 300 times and then ran the script in a console window and then processed the 300 downloaded files with KEDIT. No use of GUIs. For such work, usually GUIs are useless.

=== Files

I'm totally in love with Microsoft's NTFS (new technology file system or some such) file system. I would like better documentation on the file and directory (I HATE Apple introducing the word folder) attributes such as system, hidden, archive, etc. and on locking and concurrency.

E.g., for handling those 300 files, do that in a subdirectory -- don't let other files get in the way and can copy, delete, etc. easily.

=== Manipulating the Text

Since I work with text, I need good tools for handling text, and my most important computing tools are the text editor KEDIT, its macro language KEXX, the scripting language REXX, the D. Knuth mathematical word processing TeX (I write TeX but no LaTeX), a spell checker ASPELL that comes with the TeX distribution I use, the .NET languages and object library, for some important work an old Watcom Fortran compiler (with the very nice IBM linear programming package OSL, optimization subroutine library).

So, I want good tools and if necessary write my own and don't want little apps doing things for and/or to me. Really, so far I have no apps at all.

=== Version of Windows?

Since I like text and console windows so much, the stuff Microsoft added to Windows 7 Professional to get to Window 10 Home Edition I don't want. I just like that version of Windows 7. I can think of some improvements I would like a lot, but none of those are in Windows 10.

For what Apple wants me to use, no way, not a chance, never.

=== Good Windows

To me, one of the best things about any of the versions of Windows is that they are still good to great places to run old command line software. E.g., KEDIT goes back to PC/DOS, OS/2, Windows 95, ..., Windows 7, Windows 10.

=== Typing for Developing

I used KEDIT with KEXX and REXX to develop the software for the Web site of my startup, 24,000 .NET programming language statements in 100,000 lines of typing (lots of good comments with some little KEDIT macros to ease using documentation, etc.).

Visual Studio seems to be intended to do what I do with KEDIT -- so far I prefer KEDIT to Visual Studio. E.g., Visual Studio is part of the long standing Microsoft idea of GUIs, and I just reject that as a big step backwards. I had no trouble at all using KEDIT, etc. to develop that software. The problems were, e.g., bad documentation for SQL Server that made getting a connection string a solid week of mud wrestling. Finally someone in Microsoft's SQL Server organization solved the problem.

=== Documentation or Experiments

Part of the Microsoft, Apple GUI approach is no real documentation, e.g., nothing like what was written by Mike Cowlishaw for REXX or by D. Knuth for TeX, and, instead, learn just by experimentation. I don't like that experimentation -- e.g., Windows 10 has some huge number of special keystrokes that do things, and I still have no knowledge of what those keystrokes are or do. I encounter those keystrokes by accident; some windows pop up, and I work to close them ASAP since whatever they are I know I want nothing to do with them. I don't like undocumented tools.

=== No, I Don't Want That

Generally what the Apple, Microsoft people and their app developers have in mind to please me just makes me angry. What they are offering me, I don't want. To me their work just gets in the way of my work; I hate their work and their assumptions about my work.

So, on several points I agree with the OP on "universal computing". If Microsoft will keep console windows and let me run old software, I will be happy. To make me happier, they can do better on documentation and tools for common tasks in system management. E.g., I'd like better means of backup and restore. For more, they can have fewer bugs and security problems. For more, I would like some good documentation for Microsoft's Power Shell. For the rest of the industry, I wish I could get a keyboard as good as IBM shipped with their PC/AT.

The stuff in this post reads straight out of the late 90's/early 2000s.

The Banach-Tarski paradox is fine with me: Not all subsets of the real line or R^n for the set of real numbers R and positive integer n are measurable. Yes, the usual example of a non-measurable set uses the axiom of choice. The sets in the Banach-Tarski paradox are not measurable -- okay.

Of course the natural numbers can be put into 1-1 correspondence with the rationals -- how to show that is the classic Cantor diagonal argument.

At a tea before a seminar, I asked Max Zorn what Paul Cohen had just proved. Zorn didn't explain it and instead just loaned me his copy of Cohen's paper -- I lost it in a move! If Zorn wasn't strongly interested in Cohen's proof, then neither should I be.

For any set of axioms, there will be statements we can neither prove nor disprove. Surprising, interesting, got it.

Zermelo-Fraenkel set theory with the axiom of choice seem fine to me -- now back to the real work!

WOW! Follow the link of the OP and see a lot of colorful Romanian culture!

So, they make haystacks to feed the livestock in the winter. Yup, that is a near universal need for any farming with animals, that is, a near universal foundation of the economy of the world.

Sooo, also in the US, we have to harvest and store hay for the winter. So, now I remember: As a child, late one summer our family took a vacation and visited the area where Dad grew up, in the small town of West Valley, NY, a little south of Buffalo, NY. Dad's family didn't farm: His father owned the West Valley general store, and his step father owned the West Valley feed and grain mill. But one of Dad's childhood friends ran a dairy farm, and I got to visit, right, just as they were "making hay while the sun shined".

So, I rode in the wagon back to the barn: Now the situation is different from Romania and a lot more from Europe can see with a simple Google search. The barn was, yes, for the animals. But the barn had a second floor, the hay loft -- right, another solution to the same problem as the Romanian haystacks.

The hayloft was a large flat area, a whole second floor. At one end of the barn, there was an opening in the side wall. And there was a pulley centered in the opening, maybe 8' above the floor of the hayloft, and maybe 4' outside of the end wall of the barn.

Now want to move the hay to the hayloft, a way to store the hay for the winter and an alternative to the Romanian haystacks. So, grab maybe 200 pounds of the hay with some curved steel fingers like a hand grabbing a lot of spinach, run a rope from the top of the fingers to the pulley and from the pulley to a heavy horse, have the horse walk slowly and pull on the rope, and, thus, move the hay up to the opening on the side of the barn. Then pull the hay into the hayloft of the barn. For more, there can be a trolley system that will move the hay well into the barn and let it fall onto the floor of the hayloft.

So, right, "make hay while the sun shines" so that avoid the hay being wet and rotting -- same problem the Romanian haystacks solved.

Another step in handling hay can be to make hay bails. Some hay bails are cylinders maybe 6' in diameter and 5' high. Maybe a more common hay bail is rectangular -- there is a really good video of making such hay bails in


At one point the video shows that the hay bails are held together with wire. That would be the universal repair material, along with chewing gum, called bailing wire.

Now I more fully understand what I saw in that West Valley vacation when I was a child!

To continue, following the URL of the OP here, can see a lot of Romanian farm houses. They don't look like they have Internet, TV, telephone, or electric power.

Not so long ago in the US: When my future wife invited me to her home, they had a working farm. In one respect, they were modern in that both parents had college educations. But they had no cable TV, and their telephone was only on a party line. The area had only recently gotten electric power. Since then that area has become fully modern.

It does appear that in technology (A) in some ways within a few decades US farming areas were only a few decades ahead of Romanian farming areas but (B) in the last few decades the US farming areas have moved ahead much faster than the Romanian ones.

Then, it is not clear that the US farming area people are happier than the Romanian ones!

I would like to say, having lived in villages very similar to these in Transilvania, Romanian farm houses absolutely have internet, smart phones and internet.

WOW. Thanks. The images I saw showed the houses as nearly all log cabins where the logs were not painted, and I didn't see anything that looked like electric power lines. Can get Internet from just cell phones although of course need a source of electric power to charge the smartphones.

Maybe those areas of Europe are catching up, in the technology that is worth having, with England, the US, etc. I wish them well and hope so.

If you look at TFA there's a photo of a haystack in front of a small manor sized brick house :)

The photos of traditional wooden cabins are more for tourists, or in very remote areas.

Edit: come to think of it, wooden cabins aren't so traditional outside the mountains.


This looks more like the traditional poor peasant's houses (photos from 1933) and they're made of mud bricks (not sure what the english term actually is) with some wood reinforcement in the walls.

There's more on how to harvest, chop, store for the winter, and feed cattle in


The end of this video shows feeding the silage to a herd of cows.


No use of barns or hay lofts!

Store the silage in silos or on the ground covered with plastic.

Scale is nicely large: In total for the one farm, to store on the ground, cover an area maybe the size of a football field, 100 yards by 40 yards, maybe 10 feet deep.

All of that is in productivity per person way above what I saw as a child or young adult in visits to US farms.

"TFA" means WHAT the heck?

I just followed the URL given in this thread and commented on what I saw in LOTS, dozens, of pictures and compared with some of what I know about how the US has solved the problem of gathering and safely storing summer hay for winter animal feed.

In the URLs I followed, nearly all the houses I saw were what in the US are called log cabins and had old logs and no paint.

I didn't see any "mud bricks".

From what I saw, there was no sign of electric power. Thus there would be no "land line" telephones either; from that I concluded no "telephones" -- I didn't think of the old satellite phones, cell phones, or the new versions of phone and internet via constellations of satellites.

Maybe everything I saw was just for tourists. Okay. I saw a LOT of pictures; maybe they all were for show.

In that case, John Deere should open some branch offices (if they haven't already) and sell some of their terrific, highly computer controlled, automated farm equipment, in particular for converting hay fields into bails of hay. Or for planting, growing, and harvesting wheat, corn, or soy beans.

But John Deere can change the culture of a farming community: Their equipment can be highly productive but also highly expensive. The pair can mean that a father and his two sons can run a farm of 5000+ acres. E.g., my father in law's farm was just 88 acres. He raised a family on that (40,000 chickens per batch) and got all three of his children through college. And as electric power came along, he led in the effort and became the head of the local electric utility and lobbied in Washington, DC for the relevant legislation (REMC -- rural electric membership cooperative). He did well: One of his daughters was Valedictorian in high school and Phi Beta Kappa, Summa Cum Laude, Woodrow Wilson Fellow, NSF Fellow in college, and high end research university Ph.D. One of his grand daughters was Valedictorian in high school, Phi Beta Kappa in college, got her law degree at Harvard, started as a lawyer at the high end New York City law firm Cravath-Swaine, got an MD, and now is practicing medicine.

But the 88 acres doesn't work very well now; what he did was tough; it would tougher now. Due to progress in productivity such as from John-Deere, there has been massive consolidation of those little farms into farms of a few thousand acres operated by remarkably few people, for growing wheat, corn, soy beans, or grazing cattle for milk or meat.

So, maybe Romania is doing the same, the hay stacks put together by hand with wooden tools are just for the tourists, and John Deere is getting a lot of business. Good for Romania.

> From what I saw, there was no sign of electric power.

You should consider looking again, I count electric power lines in no fewer than 6 of those pictures.

No way are we looking at the same pictures.

First one with power lines is the 11th one down, "Image Credit Flickr User Camelia TWU".

And in #16 there's https://www.flickr.com/photos/horiavarlan/4884236029/in/phot... and those sure do look like power lines to me.

Hay lofts are common in România too. My grandma had hay stacks for long term storage, which would be used to refill the hay loft as it emptied. A hay loft on its own can’t hold enough hay for an entire winter.

Uh, depends on the size of the hay loft! The barn with hay loft I saw growing up seems to have been good for the whole winter, and that was on a dairy farm. Another approach on such a farm was a silo, a tall cylinder.

It's "bale" not "bail".

1987? How could the poor students have known?

Gee, 1987, I was at IBM's Watson lab as a researcher in their effort at the time in artificial intelligence.

So, at the lab we had an ambitious computer services group, and they had for our general use six mainframes, I believe all single processor machines, running the operating system software Virtual Machine/Conversational Monitor System (VM/CMS).

For access from home, I had a PC/XT with some special software to use a standard dial-up land line telephone line to access the mainframes. As I recall, the data rate was about 30 characters per second.

So, the mainframes were likely water cooled. As I recall, the last of the water cooled mainframes had a clock speed of 153 MHz.

In contrast, the last desktop I plugged together has an AMD (American Micro Devices) processor, FX-8350. It has 64 bit addressing and 8 cores. The standard clock speed is 4.0 GHz. To buy the processor, at Amazon I paid their usual price, quantity 1, ~$100. Tiny little thing; came in a tiny little box. The processor has been fine: It is the basis of my startup.

So, let's compare, counting clock ticks:

The 6 mainframes had best I can guess at most in total

6 * 153 = 918

million ticks per second. The AMD has

8 * 4.0 * 1,000 = 32,000

million ticks per second.

Then we can take the ratio

(8 * 4.0 * 1,000) / (6 * 153) = 34.858,387,799,6

Call it 35. Soooo, counting clock ticks my desktop with the $100 processor from the tiny box is 35 times faster than all six of those mainframes combined.

With speculative execution, out of order execution, pipelining, etc., there is on average some number of instructions per clock tick. So, first cut here, and maybe a gift to the IBM mainframes, for the two processors count the average number of instructions per clock tick as the same.

So, this fantastic, we're talking not just world class but unique in civilization class, improvement in price/performance, is much of what enabled Google, Facebook, Amazon, Apple, Microsoft, AMD, TSMC (Taiwan Semi-Conductor Manufacturing Corporation), smart phones, solid state disks, and essentially all of the Internet.

E.g., for communications data rates for connect from home, now with 1 Gbps Internet versus then, 1987, 300 bps, ..., to the 56 Kbps of V.90 of 1995.

So, how the heck in 1987 was a computer science student or early career person to anticipate and actually believe in this fantastic, all-time unique step up in price/performance of computing and communications along with associated infrastructure?

This progress blows away wood, wheels, stone, copper, iron, steel, steam, electricity, and broadcast radio and TV.

Now, what are we going to do with this suddenly cheap, powerful computing, communications, and infrastructure?

Didn't read every word of that, but it seemed a particularly long-winded version of Moore's Law.

It appears to me that Graham's essay is missing information and emphasis on one more crucial input to doing "great work" or being successful at all. That input is the importance of and good approaches to

===>>> Good Problem Selection <<<===

including good initial problem selection.


(1) By accident, I had a relatively good career going around DC in applied math and computing for US national security. I thought it would help my career to know more math so got a high quality Ph.D. in applied math. I WAS then better qualified, but I also had broken my career continuity. My brother saw that early on: "After your four years in graduate school, you will be four years behind in your career."

And, I didn't realize that the opportunities in DC were unique or nearly so. I would, could, should have (A) not assumed my career would work away from DC and (B) stayed in DC.

Lesson: If you have a good career going in DC, NYC, Silicon Valley, wherever, don't assume without good evidence that that career will also work in another location.

(2) Near the end of my Ph.D. work, I had a part-time job and did some good things. Then occasionally a person never seen before was in the office for a few hours one or two days a week and wanted to chat with me about whatever, even sports. Eventually he got my opinion on what the US did in Viet Nam. In retrospect I was being recruited for some high end highly secure slot but violated a rule Mom gave me -- never talk about politics, sex, or religion. I had talked about politics. Don't do that. So, being frank and honest about Viet Nam cost me a good career step.

(3) Dad's idea about a career was to get a good education and, then, a good job based on that education. Eventually I concluded that, nope, instead, with or without a good education, it is from important to crucial to own a business and to make it successful, and early in a career near the top of the TODO list is to find a way to own and run a successful business.

(4) I didn't understand an important point about career dynamics: (A) It is tempting to guess that academics is good preparation, maybe the best, for a non-academic career. (B) Instead, for a successful non-academic career, it's important to ride the waves of the hot topics, and, bluntly, academics is largely out of touch with what the hot topics are. (C) While some academics can help, possibly a lot, there is a special way to know what the hot topics are in a non-academic career -- see what work and topics people are getting hired for.

Net, be reluctant to let academics tell you what will be important in a non-academic career and, instead, let a non-academic career tell you what topics are important for you in academics. And if have a good ugrad education, then pursue those topics while on the job by independent study evenings and weekends, maybe partly on the job -- to pursue those topics, don't leave the job to become a full time student.

E.g., when I was in my Ph.D. program, there was a prof, soon famous, who apparently had been hired as a consultant to work on a topic, call it, X. Well topic X was in practical computing, a bit tricky, not at all a topic in academics, and from my non-academic work a topic I knew well. With resentment the prof asked me: "I suppose you know X?" I had a one word answer: "Sure". Really it was a collection of topics such as X that had my career going well, e.g., annual salary six times the cost of a new high end Camaro. For that summer job working on topic X, the prof was about to have a tough time!

(5) One possibly good career direction is to get a job in a startup in its early days and get stock. For the stock, before joining (A) get the stock deal in WRITING, (B) study the details of the stock deal (e.g., vesting), and (C) have the deal reviewed by an appropriate lawyer. Handshake deal? Don't do that.

(6) Applied math and, usually, computing work or don't independent of human emotions. People, however, commonly are driven more by their emotions than what is true in math or computing. It is crucial, then, to pay close attention to the emotions of people.

As a special case, one way to get people to hate you and work against you is to present some work in math or computing that (A) has power important for the organization, (B) is understood only by you, and (C) looks rock solidly correct and, thus, not easily ignored. Then the people who hate you maybe cannot find fault with your work but might be able to arrange to get you fired.

Part of one way to get you fired is gossip -- be aware that you may have to defend yourself proactively, guess what you are secretly accused of and then defend yourself against the accusations.

More generally, whenever you start to be especially successful, you will attract attacks like some piles attract flies.

(7) There is discrimination in the US. Not often mentioned is that there are some cases of severe discrimination in the US against native born Christian males of European descent. So, it can be important to realize that fact.

(8) Try to keep confidential and not to report on work in progress: If your work looks good, your enemies will try to kill off your work before it starts to show its power or arrange to take credit.

(9) There can be corruption, e.g., people getting quite secret financial kickbacks. So, you can get a bitter enemy if, even without realizing it, you threaten their corrupt money. So, at least try to guess who might be corrupt and how, and then try to avoid that person or their source of corruption.

Of course, if there is someone corrupt in the part of the organization chart you manage, you will likely want to get this person moved out. Of course if the corrupt person has some close, say, family, connection with, say, a member of your BoD, then you have to be careful!

Point: Be aware that corruption exists.

NYS crowded, right? Well, just look at a map, e.g., via Google maps. Can look for lots of wilderness places in NYS, but, to make looking easy, just go to the Catskills or, better yet, north of Albany. There are huge areas of forest. For a name, try the Adirondacks. Somewhere up there they had a Winter Olympics -- plenty of ice, snow, and skiing!

Recently I escaped from NYS and went to east TN. There have the Smoky Mountains with huge areas of forest and hills.

Long lived in Maryland and often went to the mountains of VA -- huge areas of meadows, forests, deer, bears, etc., that is, Shenandoah!

And that's all well east of the Mississippi River.

Yes, technically, "Recreational land is much more sparse east of the Mississippi." but still huge. Yes, in Montana, Wyoming, Colorado, Idaho, Nebraska, it's harder to find people or buildings and easier to find wilderness!!!

But, sure, if go to Canada and north a few hundred miles of the border with the US, then find very sparse areas -- it seems that most of the people in Canada live not so far from the US.

But for the US again, if want wilderness, try Alaska. Of course we don't hear so much about the wildest parts of Alaska -- maybe the reason is so few people come back alive!

There is a point rarely mentioned but likely important: Academic research is highly competitive. Doing well in an academic career is essentially the same as doing well in the competition.

In particular, the Federal government, Congress, the NSF and NIH don't have to check the quality of the work closely for each grant because the academic system will check the quality before a researcher gets progress in their career.

Applications are open for YC Winter 2022

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact