HNHacker News
TopNewBestAskShowJobs

greggyb

3,175 karma · joined December 23, 2014

hn (at) <username> (dot) com

Data intensive applications and analytics.

Functional programming.

Systems design.

Consulting.

submissionscomments
greggyb··on Breaking Up with Long Tasks or: how I learned to group loops and wield the yield
I’ve come across that as well, but in presentations. I think it is either Alan Kay or Bob Martin who likes to use the statistic. I recall it being framed in terms of the growth of tech/programming as a profession: it’s doubled every 5 years; the fact that 50% of developers have less than 5 years of experience. This is simply exponential growth stated in two different ways.

I do a lot of work on a platform that has doubled in user base yearly since its inception, for about a decade (though I think we are at an inflection point now or soon), and it is wild to have “experts” be those with 2 or 3 years of experience. Having used the platform for 11 years, now it is crazy to believe I have more experience than 99.9% of the field.

greggyb··on China is the manufacturing superpower
I imagine most courts in the US would, in fact, find “3N” as blatantly infringing the 3M trademark.
greggyb··on China is the manufacturing superpower
The price (that consumers are willing to pay) is the only decent proxy for quality we have. Quality, of course, includes consumers’ preferences.

If the one car sells at a market price of 1/5 of another, then the only reasonable conclusion is that, as judged by consumers, one car is worth 5x as much as the other, and therefore is 5x better.

That’s what market prices are: the collective valuation of goods/services and their alternatives. Of course, the market clears at a price that is a combination (you can think of “average”, but that arithmetic is not the process by which prices are discovered) of the preferences of all consumers in the market.

If you have a preference that goes counter to the market, e.g., if you do not think those two cars are 5x different in quality, then there is a great deal for you!

If you think the worse car is only 2/5 the quality (rather than 1/5), then you can get something for half the price you’d be willing to pay if the rest of the market agreed with your preference.

If, on the other hand, you think the better car is 10x the quality of the worse car, then the same “deal” is true for you. You’d be willing to pay 10x the cost, but you only have to pay 5x the cost.

There are plenty of good macro and micro econ courses available for free if you’d like to dig into this sort of thing more.

greggyb··on China is the manufacturing superpower
I don’t think that you’re correct about US O&G revenue.

The US produces 13M barrels of crude and condensate per day: https://www.eia.gov/todayinenergy/detail.php?id=61545

The lowest recent spot price is $69.5: https://www.eia.gov/dnav/pet/PET_PRI_SPT_S1_D.htm

This tells you that annual production of just crude oil is > $300B/yr. The US also has plenty of downstream processing after production (https://www.eia.gov/dnav/pet/pet_pnp_unc_dcu_nus_m.htm), which means that total annual revenue must be significantly greater than this.

greggyb··on Phase behavior of Cacio and Pepe sauce
Stir.
greggyb··on Databases in 2024: A Year in Review
I'll probably come across as a shill here, but there is a lot going on with SQL Server, all included in your license (Standard Edition has limitations on scaling).

Some of these things are merely passable, some are great, but it's all included. The key takeaway is that SQL Server is a full data platform, not just an RDBMS.

- RDBMS: very solid, competitive in features - In-memory OLTP: (really a marketing name for a whole raft of algorithmic and architectural optimization) can support throughput that is an order of magnitude higher - OLAP: Columnstore index in RDBMS, can support pure DW style workload or OLAP on transactional data for near-real-time analytics - OLAP: SSAS: two different best-in-class OLAP engines: Multidimensional and Tabular for high-concurrency low-latency reporting/analytics query workloads - SSIS: passable only, but tightly integrated ETL tool; admittedly in maintenance mode - SSRS: dependable paginated / pixel-perfect reporting tool; similar to other offerings in this space - Native replication / HA / DR (one of the only things actually gated behind Enterprise) - Data virtualization: PolyBase

If you're just looking for a standard RDBMS, then there's little to justify the price tag for SQL Server. If you want to get value for money, you take advantage of the tight integration of many features.

There is value for having these things just work out of the box. How much value is up to you and your use cases.

greggyb··on Load is not what you should balance: Introducing Prequal
It's not weird, but incomplete. It is broadly captured in the Economics concept of "willingness to pay", loosely the maximum price someone or some firm would be willing to pay for something of benefit to them.

This contrasts with "willingness to accept", loosely the minimum compensation someone or some firm would accept to produce a good or service (or accept some negative thing).

Neither of these is sufficient to determine the price of something precisely, but, in aggregate, these concepts bound the market price for some good or service.

greggyb··on Qutebrowser: A keyboard-driven, Vim-like browser
If you are looking at a link you want to follow, you press a single hotkey (default 'f'), and just type the two letters that appear directly where you're looking. If you want to open a lot of links in the background, e.g. if you're looking at the front page of HN, you can use the rapid follow action (default ';r') and just type a couple letters for each link you'd like to follow.

Personally, this is much easier on my RSI than mousing to each link. Especially on a laptop with a trackpad.

Additionally, the browser is fully configurable in Python, to a much greater degree than what you get from Firefox or Chrome.

One seemingly small thing that I find quite convenient is that I can use reader mode or open a PDF and still keep hjkl for scrolling. I also use Tridactyl in Firefox, but Firefox's reader mode and opening PDFs turns disable extensions in the tab, so you can't use these scrolling keys.

As for next and previous page, I think you're referring to history navigation forward and back. Qutebrowser includes built-in functionality to navigate in multi-page documents to the next page and previous page, e.g. documentation sites or html books. That shortcut is '[[' for previous and ']]' for next. So, for example, if I am on this section of SICP: https://sarabander.github.io/sicp/html/1_002e3.xhtml#g_t1_00... and I press ']]' I immediately go to https://sarabander.github.io/sicp/html/Chapter-2.xhtml#Chapt... . This happens regardless of browser history. "Back" and "Forward" are different from "Previous" and "Next". I do not know of a similar piece of functionality in other mainstream browsers.

This is all highly opinionated. I like vim-style shortcuts. I vastly prefer the experience of using qutebrowser (or a vim-like extension for another browser) over using a mouse exclusively or a mouse and keyboard mixed.

If you, or someone else, prefers a different style of interaction with the browser, that's fine. It's not about right or wrong or better and worse, at least not in the abstract. It's about a user interaction paradigm that works well for an individual.

greggyb··on Every V4 UUID
Thanks. Now I can always double check mine!
greggyb··on Every V4 UUID
I had a database of all SSNs for a while, but it was on a work laptop, so I didn't get to keep it.
greggyb··on Lossless Log Aggregation – Reduce Log Volume by 99% Without Dropping Data
We have a process monitor that basically polls ps output and writes it to JSON. We see ~30:1 compression using zstd on a ZFS dataset that stores these logs.

I laugh every time I see it.

greggyb··on The Curse of Markdown
The text graying out and scrolling effect leads to overlaps for me. Firefox 133.

https://i.ibb.co/NryLFgJ/codehike-overlap.jpg

greggyb··on When was the famous "sudo warning" introduced? (2019)
Any reason not to run SSH on localhost only? This way there would be no network requirement whatsoever.

I am not trying to criticize, but am just curious what you gain by having this access split across two machines?

greggyb··on Way too many ways to wait on a child process with a timeout
Not so much about timeouts, but related in that it is based around managing children processes:

The lineage of tools descending from daemontools for service management is worth exploring:

daemontools: http://cr.yp.to/daemontools.html

runit: https://smarden.org/runit/

s6: https://skarnet.org/software/s6/

dinit: https://davmac.org/projects/dinit/

greggyb··on How do you deploy in 10 seconds?
Why did this get flagged? Seems very on topic for HN and is not blogspam or clickbait.
greggyb··on Steve Ballmer was an underrated CEO
> For example everyone agrees that Intel is a failed company, but if you actually look at them they still have majority marketshare of x86 CPUs in both client and server. So would you say that they successfully defended their traditional business or would you call them a failure?

I do not think we live in the same world. As one data point, the majority of analysts still list Intel as a buy or a hold.[0] In the very same industry, AMD showed us that it can take more than a decade to turn around a poor architectural turn in semiconductor manufacturing. Their stock price peaked at the beginning of 2006 and then were in the doldrums until 2017-2018 (depending when you want to count their return). It may well turn out that we are witnessing Intel's death spiral, but to claim that with certainty, and further to ascribe that belief to "everyone" tells me that we see the world incredibly differently.

[0] https://finance.yahoo.com/quote/INTC/analysis/

greggyb··on Steve Ballmer was an underrated CEO
Windows runs in more places than it probably should. And the OS is just fine on other platforms. Windows on ARM works quite well. As I understand it, it worked just fine on Itanium. NT launched with support for PowerPC. I'm sure RISC-V support will be there pretty quickly. NT was designed for this sort of portability.

Most user software is not ready for the transition. Microsoft doesn't impose the same kind of control over its application ecosystem as Apple does. This prevents Microsoft from being able to make architectural shifts like Apple has. Arguably, this more open and flexible ecosystem (and bending over backward compatibility) has been one of the driving factors in Windows's popularity over MacOS. I have no strong opinion about which is a better business strategy.

greggyb··on Steve Ballmer was an underrated CEO
AWS was the first mover, and they execute very well on products. Despite this, AWS market share has been pretty flat for a while.

Azure was a second mover. Microsoft executes on platform and partner ecosystem very well. Azure is still growing market share pretty quickly.

greggyb··on Steve Ballmer was an underrated CEO
> Basically the cloud for everyone but the tech companies

And most companies are not tech companies. This is something that tends to be lost in HN discussions (not saying that applies to you, specifically).

I've spent a lot of time in the Microsoft world. I worked for AWS as well. In general, Microsoft executes on platform and ecosystem in a way that works very well for a lot of customers. In general, AWS executes on products better, which tends to appeal more to those who are focused on technology, specifically.

greggyb··on Steve Ballmer was an underrated CEO
See my sibling comment. Windows Server dominates server OS revenue share at ~70%. They get most of the money that people are willing to spend on server OSes.
greggyb··on Steve Ballmer was an underrated CEO
> I don't buy the "MS purposefully let IE die to avoid anti-trust" story

I don't know who you'd buy it from, certainly not me. If that's what you've read into my comment, that is reflective of your perspective, not mine. IE was never that good. It was bundled with the predominant operating system, so had massive market share. As soon as Microsoft had to open the OS more and draw a line between OS and IE, it was pretty much inevitable that there would be other browsers. And too-aggressive positioning of IE would be highly legible to regulators.

I agree that, absent regulatory scrutiny, Microsoft could have flexed muscle here to effect different outcomes in the general internet space. I disagree that such a decision would be a good idea for Microsoft in the climate of the aughts.

Server market share is tougher to find, but it appears that Microsoft captures a large majority of revenue in the server OS space. I know the install base proportion is smaller than the revenue proportion, but it seems to me that they are capturing most of the market willing to pay for a server OS.[0]

As to your 7 major battlefields and Microsoft's major advantages:

1. Operating systems. Most other major OS vendors collapsed or were collapsing in the 90s. It seems naive to expect that there would continue to be no pressure on the OS front. Part of their dominance was circumstantial. Nevertheless, they continued to dominate desktop market share and server market share (both by revenue). Losing install share to OSes that people don't pay for is expected, especially as those options improve.

Apple came over to x86, making MacOS (again, an OS that people don't pay directly for) much more viable, because application developers no longer had to target two ISAs.

And mobile came to the fore. Consumer devices is a separate category, though.

Overall, I'd consider this a win for Microsoft and a continuing area of dominance, in the face of many new entrants and threats: ChromeOS; tablets becoming a viable option; Linux becoming usable for laypeople; and much of computing moving to the browser such that the OS doesn't matter as much, yet Windows still dominates when OS doesn't matter.

2. Search. Google was already positioned for market dominance in the early 2000s. Microsoft began their indexing project in the early 2000s. Other players entered and left the market. Amazon had a search engine, and they are all about the consumer market (especially before AWS). That folded in the mid aughts. Yahoo collapsed. AOL, Excite, Lycos, and many others failed. Yahoo now uses Bing's index. DuckDuckGo uses Bing's index.

Bing is a solid #2 in search, earning $10Bs annually. That is an entire business.

Microsoft started from a disadvantage on search and is #2. #2 is not a loss. Also definitely not an overwhelming victory, but the world is not binary. Writing off a business with $10Bs of revenue as a loss seems absurd ... may we all be so blessed as to lose like this.

3. E-commerce. Amazon was already positioned for dominance and was solely focused on this for the early aughts, and primarily focused on this for the duration of the aughts. Microsoft had no major presence here.

To make significant headway, Microsoft would have had to go against one of the most single-minded and aggressive companies that was out there. (Amazon is less single-minded today). And we know what Amazon's successful strategy is in retail, and that is essentially no profit. Microsoft, both in terms of its corporate culture and in terms of its major shareholders, would never have been able to compete with Amazon in that way.

Microsoft had no major advantage here. This is certainly not a win, but they didn't want a piece of it anyway.

4. Social. Do we count messengers as social media? If yes, then yeah Microsoft had MSN messenger. If we don't count messaging, then they had no specific expertise or advantage here. In fact, many of their attempts at more lighthearted computing were panned: clippy, Microsoft Bob, Comic Sans. Microsoft has always been nerdy and never cool until others made nerds cool.

Regardless, they purchased Skype and LinkedIn and have a combined MAU of ~1.2B, putting them comfortably in the top 10 of social media companies.[1] I'll note that this list includes Teams, but I'm going to count that as enterprise software, because it is included with almost every O365 SKU.

LinkedIn was a Satya purchase, so Ballmer can't get credit there. Nevertheless, I wouldn't argue that Microsoft had a major advantage for social media. I'd agree that they lost in this space.

Arguably you could put github in here, too, but also a Satya acquisition.

5. Consumer devices. Microsoft is a software company to its bones. Windows was a definite advantage here. But Microsoft does not have a strong history of device manufacture to lean on. They tried to acquire for this with Nokia. And yeah, they failed here.

Again, though, they were against one of the most single-minded and aggressive companies out there, this time Apple. And Apple already had the pedigree for prestige consumer devices. Apple was defining hip and cool with regard to technology through the aughts. Again, Microsoft is neither.

Additionally, they were up against another competitor for whom smart phone market share was an existential concern, Google. Google does not have the single-minded focus of Apple, but when it comes to threats to their core business (selling ads), they can execute well enough.

Microsoft does not have the single-minded focus of Apple, nor did they have the fundamental imperative to make something work on mobile that Google had. Still, Microsoft should have done better here, and in fact Windows Phone OS and devices were largely well reviewed. Nevertheless, they couldn't make headway against their rivals.

Microsoft had an advantage here and lost.

6. Enterprise software. Microsoft was the dominant player and remains the dominant player in this space. O365 is undisputed. You will pry Excel from many people's cold, dead hands. Microsoft may not be cool, but no other tech company has a product on ESPN. The Excel championship is something else.

And I will note, specifically, that O365 began under Ballmer.

Microsoft is also a huge player in data platforms. They have specific product wins and losses, but as a whole, they do very well in this space. Data is a specific niche of mine, and I can tell you that there is no other vendor than Microsoft that companies will routinely go all-in on for their data platform. Don't get me wrong, this is not the norm, but I only ever see it happen on Microsoft, never for another vendor's platform all-in.

I covered Windows Server above, but again I will note that they capture a large majority of server OS revenue share (which I argue is the category to measure this in).

Microsoft is a dominant player in this space, and I would count this as a win. It was part of their core business, and that business tripled sales and doubled profit under Ballmer.

7. Cloud. Amazon moved first. Microsoft followed quickly. They are #2 in the market and growing fast. Google is no threat to either Azure or AWS. Alibaba probably will be in the next 10 years.

Azure would not exist in the way it does today without significant investment under Ballmer.

#2 for a second mover (using "second" as "non-first", here) is not at all a loss. And threatening to overtake the first mover is a powerful position to be in.

I expect you'll disagree that any #2 position is worthwhile, or at least that seems to be your stance.

Regardless, I'd put the counts as:

Advantage: 4: OS, Enterprise software, cloud, consumer devices

Disadvantage: Search, E-commerce, Social

Competed: 6: OS, Enterprise software, cloud, consumer devices, search, social (questionable)

Success: 4/5: OS, Enterprise software, Cloud, Search, Social (social only by acquisition, mostly post-Ballmer, but they were in a position to make multi-billion dollar acquisitions, so Ballmer can't have ruined things)

Failure: 1/2: Consumer devices, Social (if you want to argue that only Satya's acquisitions count, and no credit toward Ballmer).

Did not compete: E-commerce

[0] https://www.statista.com/statistics/915085/global-server-sha...

[1] https://en.wikipedia.org/wiki/List_of_social_platforms_with_...

greggyb··on Steve Ballmer was an underrated CEO
> MS should've been able to simply just extend their OS monopoly into all platforms and all architectures, but they didn't, and to a vast swath of the world have become irrelevant, and worse, have lost their ability to become relevant.

Microsoft is, pretty famously, on the receiving end of one of the most significant antitrust judgments in modern history. Choosing to further a monopoly seems that it would be a phenomenally bad decision for the company.

Despite that, Windows remains the dominant operating system for businesses worldwide. So I would argue that the OS is far from "irrelevant".

Beyond the OS, they are comfortably #2 in the public cloud market, with little threat from #3. Indeed, #1 has been relatively stagnant in market share, while Azure has been steadily growing.[0] It seems that a consistently growing market share in such a large industry shows that not only are they relevant, but they are becoming more so, and have not "lost their ability to become relevant". Additionally, it seems that they are "overcoming competition".

> It's a decline from being the monopolist to simply a player ....

> MS should've been able to keep control of the internet, but they lost their moat to Google (Chrome),

They legally could not maintain that monopoly. Again, see the antitrust ruling. The antitrust case was about the impact on Netscape, and was too late to save Netscape. But it is a pretty straight line from a case about bundling IE with the OS.

To be clear, the finding in this case originally held that Microsoft needed to be broken up.[1] Microsoft won on appeal, because of impropriety by the original judge in the case, but the appeals court upheld all findings of fact.[2]

Much of what you are saying Microsoft should have been able to do on the basis of their OS monopoly would have been begging for further antitrust action.

> Bing was a decent win

Bing, on its own, would be the 12th largest tech company in the world, per the original article.

And Microsoft is worth $3T today, largely on the basis of investment under Ballmer (and continued strong execution under Satya). Is the argument that Microsoft should instead be a $10T or $100T market cap company today if you graded Ballmer better than a D?

[0] https://www.statista.com/statistics/967365/worldwide-cloud-i...

[1] https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor...

[2] https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor...

greggyb··on Steve Ballmer was an underrated CEO
If you trust the article, then Azure and O365 are each, independently, easily Fortune 100 companies if separated. These "couple of wins along with many many losses" are some of the most valuable products in the world.

Imagine a VC fund that invested in a few dozen product companies, two of which were Azure and O365. Is that a 6/10 VC company? Why is the logic different for a CEO making bets for a company's next several decades?

greggyb··on Steve Ballmer was an underrated CEO
Bing Ads is big business. Digital marketing is enormous. Google and Facebook have larger portions of the pie, but a sliver of a huge pie is still a lot of pie.
greggyb··on Winamp deletes entire GitHub source code repo after a rocky few weeks
> And this image does not depict a pipe

In fact, it does. Literally, "This is not a pipe," because an image of a pipe is not a pipe.

From the artist:

> The famous pipe. How people reproached me for it! And yet, could you stuff my pipe? No, it's just a representation, is it not? So if I had written on my picture "This is a pipe", I'd have been lying!

Sourced from Wikipedia: https://en.wikipedia.org/wiki/The_Treachery_of_Images

greggyb··on Net 9.0 LINQ Performance Improvements
Units of measure in F# are useful for more than unit conversion workloads.

They are compile-time type information with no runtime overhead. So if you're using a lot of interrelated arrays, you can make a unit of measure for indices into just a specific type of array. This allows type-level enforcement of never using the wrong index for the wrong array.

Anywhere you might want to keep categories (not category theory) of numeric values distinct from one another without intermixing, units of measure in F# can cover you.

greggyb··on The quiet art of attention
I have some graduated resistance hand exercise balls and bands for various exercise (different variations on squeezing, mobility, and extending fingers). I stare out the window at things far away while doing hand exercises.

I did pick the times because of the symmetry of the numbers (:

greggyb··on The quiet art of attention
I suffer from RSI and definitely do not move enough when I am invested in some piece of work (personal or professional). I recently installed workrave[0] and have noticed marked improvements in just a couple weeks of actually taking breaks when it indicates.

I take a 25 second break every 5 minutes, and use this time to do one hand and wrist exercise (I keep some resistance bands and hand exercise balls at the desk). I take a 5 minute break every 25 minutes. I will either do some stretches, or a quick chore (e.g. vacuum one room).

https://workrave.org/

greggyb··on The quiet art of attention
I will note that you are not responding to what you have quoted, but to an extreme re-interpretation.

The OP says "exert some mastery over", which is a far cry from the "full control" you say some people cannot have.

greggyb··on The quiet art of attention
I can recommend the Waking Up app. It has a well paced and well done introduction to mindfulness in a very similar style to the linked article. If interested, I can share a link for a free 30-day trial.

I have no affiliation and get no kickback. It has simply been quite useful for me.

← PreviousPage 3 of 34Next →