Hacker News Reader

You do, in fact, have a choice, you always did, everyone does. There are decent, completely offline solutions (Libreoffice vs GDocs), you can self-host a lot of things and do a private "cloud" on a Raspberry Pi or an old laptop (https://yunohost.org/#/ ; https://cloudron.io/ ; etc). The tradeoff is compatibility and not being trendy. However, that has always been a problem both the 80s and the 90s. Mac vs PC? Amiga vs anything?

A lot has changed since and a lot has become more comfortable. It slower, more complicated, much more annoying. But you do have the power to simply stop using whatever you dislike[1].

So: don't buy from google. Build you own cloud. If you don't like the application or if it's instable (atom), use something stable (vim?). It _will_ be uncomfortable, but it's there.

[1] I'm well aware the cost of, for example, refusing to use Facebook, but you still can refuse to use it, whereas life without WeChat in Asia is becoming near impossible.

EDIT: Sorry for possible misunderstandings, I mixed up 3 different topics in a way too short answer. My point is: there are alternatives, there are choices. You don't like the user interface of something, protest: use something else. There _are_ options, but they are not the most visible, not the most comfortable choices, and many of them will involve setting services up for yourself.

> There are decent, completely offline solutions (Libreoffice vs GDocs)

LibreOffice is slow. On a ~2010 business-class Dell laptop that still is in perfectly good physical condition, running Debian 9, with a handful of tabs in Firefox and a mosh running:

    adamantoise:~ geofft$ time libreoffice --terminate_after_init
    
    real    0m8.103s
    user    0m1.696s
    sys     0m0.520s

The rant here isn't about cloud vs. local, it's about performance and functionality. We had the cloud in the late 1980s, it was called dialing up to a mainframe, and it worked fine because software was designed to be fast in the presence of slow (by today's standards) connections. One of the examples of that -- library card catalogs running on amber-screen terminal -- was explicitly brought up in the rant.

re 1: Slow - compared to google docs? Are you serious?

EDIT: not startup time, who cares about startup time? When you load a few hundred pages doc, or a massive excel.

re 2: I only wanted to give examples, but you're right, my wording is off.

I wanted to say there are fast and good interfaces: there's vim for the terminal, there's Rainloop for webmail, miniflux for RSS reading - these are all keyboard-oriented ones, all working decently and as fast as they can. In case of web-connected ones, the speed of going back and forth is certainly an issue.

Local - as in not making connections on the network - in my opinion, will always have a speed advantage though.

re 1: Slow - compared to google docs? Are you serious?

I can't speak for the poster above yours, but I would be serious. In my experience, on the startup time, google docs way out performs open office. It's open in single digit seconds, open office is 10s+

I still prefer offline software, but the bloat can be awful. At least web tech is designed with web browsing latency expectation. I find most desktop software these days has a much larger start and use latency by far.

> re 1: Slow - compared to google docs? Are you serious?

Slow compared to decent standards of what a word processor needs. For reference, WordPad perceptually opens instantly for me, and has felt like it opened instantly since Windows 95, I'm pretty sure. I am saying that both LibreOffice and Google Docs are examples of software that take way too long to do simple things because they're busy doing complicated things that I don't actually care for. (And arguably this isn't really their fault per se, they're both designed to compete with / follow the expectations set by Microsoft Office of what an office suite is, and those expectations are that all the complicated things are available).

> Local - as in not making connections on the network - in my opinion, will always have a speed advantage though.

Depends what you're doing with the network. I do just about all of my work in the terminal with a mosh connection to a cheap server a couple hundred miles away, and I never feel perceptual slowness from my machine not being local, unless I am literally in the subway or in an elevator. And I actually gain a lot of perceptual speed because I can close my laptop and switch to my phone instantly and see the exact screen where I was.

If you're streaming binaries from the network or trying to pump animations on a 2048-by-1536 true-color retina display, yes, remote is going to be slow. Half the tweetstorm's point is that you shouldn't be designing software to need that in the first place.

If your word processing requirements are very lax you should probably just be using Markdown + an editor.

And if they're not, LaTeX is great.

> And if they're not, LaTeX is great.

And you can use pandoc to get the best of both worlds.

> When you load a few hundred pages doc, or a massive excel.

LibreOffice on even midsized spreadsheets is quite slow. Slower than opening it Google Docs in fact, which borders on shameful.

> EDIT: not startup time, who cares about startup time? When you load a few hundred pages doc, or a massive excel.

It was an example - the slowness (to me) is present for the entire interface. It was just easy to demonstrate. (Also in both LO and Google Docs I can type plain text into an existing document just fine.... I'm curious what you find slower in Google Docs than LO.)

My point is that this rant is not, as far as I can tell, a "local software good, cloud software bad" rant. There are reasons to believe that, e.g. debuggability and computing freedom, but UI speed is not one of them.

> LibreOffice is slow. On a ~2010 business-class Dell laptop that still is in perfectly good physical condition, running Debian 9, with a handful of tabs in Firefox and a mosh running:

Libreoffice is monolithically designed, as shouldn't be representative of the generic performance of modern software.

The reason is that StarOffice was envisioned as desktop environment, therefore, all the components of the suite are loaded on the first startup.

  $ libreoffice --writer &
  $ time libreoffice --calc --terminate_after_init
  
  real	0m0.104s
  user	0m0.016s
  sys	0m0.008s

If anything should be compared, then Microsoft Office should be used, as it always had a very fast startup (AFAIK, due to components preloading).

I think the article is very misguided; the problem is in the expectation that anything should be faster, even in contexts that are implicitly slower (networks, i.e. internet), or with features that will inevitably slow down the experience (e.g. browser plugins).

Using as reference the slowest software - Libreoffice, Atom, Firefox - is nothing else than cherry-picking.

In this perspective, a sibling article is correct - if you want faster alternatives, you have them.

Not everyone it prepared to host, maintain and secure a server cluster of OSS.

Having a choice is not a thing for granted. You need to work for it. You need to maintain the right to it. Those who are not prepared, learn it, or live with whatever is provided by others.

This really does smack of elitism. Not everyone simply has the time to do that.

This argument generalizes to everything, though.

Do you not like owning a car/tractor/printer running proprietary software that you're not permitted to modify? I guess you should have bought a 20 year old version and upgraded it yourself.

Do you want a choice of food that you trust won't make you sick? Then you probably should have worked for that lettuce from planting to harvest, not just complained about safety standards.

Total independence from the actions of others isn't impossible, but no one posting comments online is practicing it. And if "working for it" can include learning to use solutions other people have built, then it can also include writing this article on what's change and what's missing.

If you live in a society, talking about your preferences isn't an alternative to working for them, it's part of that work.

I merely stated you need to do, act, work for choices and for freedom. Expecting freedom of choice for granted is spoilt belief.

If you follow popular technology you get a slow stream of annoying technology thats predatory by nature. When I switched to the free software side, my experience sped up and I had finer control over everything. My files are mine and not a company's, non-free javascript is disabled so the web is faster, and my system is lightning fast since its devoid of windows telemetry. Encryption keys and offline password managers can be a hassle but the freedom overshadows the difficulty for me. Or maybe its stallman stockholm syndrome.

> I can still install Excel 97 if I want to

That's the survivor bias. There's plenty of things that you can no longer install and the things you can still install, it's mostly because there's someone that maintains their API and whatnot (thus, the company hasn't gone out of business or haven't decided not to maintains it). Look at games, GOG has an R&D department to make sure you can run the games on their platform.

I'm pretty sure there's a ton of people here that can talk about some old computer or VM that they have to keep up because of old legacy stuff.

> I mean, there is perceptible lag in character entry that never used to happen. Sure, my computer used to freeze inexplicably, but Atom crashes often enough that I don't consider that a win, and text has a noticeable delay before it appears on the screen when using Atom. Text entry is something I can implement flawlessly on a microcontroller and the majority of applications I use on a daily basis get it wrong.

Who forced you to use Atom? That's your decision to use something still heavily in development using some new technologies (not meant to be faster or more stable, just easier to develop into). It's like buying a bicycle and complaining it doesn't go as fast as the car you had before.

Exactly. When somebody uses Atom, he cannot complain about performance compared to compiled apps. And saying this as huge fan of Atom.

GOG only makes easy what the community and nerds have been doing for years. GOG made a business out of it because they pay people to "maintain" the "API". Their products are not under active maintenance nor development.

Now, if there WAS a company paying for all that maintenance it would be Microsoft, with their almost religious commitment to backwards compatibility.

> Their products are not under active maintenance nor development.

Isn't it the same? Someone that directly update the app to newer API or someone that update an API to make sure the older apps are still running is pretty much the same at the end. Someone has to do that work.

There's someone that pay for it, whether it's nerds by their time, or Microsoft by their backward compatibility support.

Even for Microsoft, at the end of the day, you pay that price through your next Windows license, and still, they may decide not to keep supporting some of it (16 bits programs are no longer supported over 64 bits for example).

It's not different from a subscription model, except that instead of paying a fortune to support most things (which may not include what you actually need), you pay only to keep support over the product you actually use and need.

I'm not saying subscription models or cloud application are amazing, but being able to keep running your old apps is not the norm and is mostly made possible because there's someone that pay for it for you. At least in a subscription model, that relation is pretty direct, you pay for the software to keep working, unlike hoping that Windows invest into that old techs to make sure they keep running.

"but being able to keep running your old apps is not the norm"

Only for the cloud cohort. Not all of us drink the kool-aid! Try living in a computing environment where the technology-rug isn't pulled out from under you every three months.

Both Windows and Linux have not had core changes to their API. The fundamentals of application programming haven't changed in decades. Why shouldn't we allow old apps to run?

I've got a good-sized collection of vintage hardware and software. Microsoft stops supporting something? All well, I've got my older versions still available. They don't get updates, but for the most part, I didn't expect them to when I bought them.

> you pay only to keep support over the product you actually use and need.

And you pay it over and over again. Until you don't need it (which might be because they removed some feature you were relying on, with no recourse).

> but being able to keep running your old apps is not the norm

Losing access to my old apps isn't the norm, unless I choose to let go of control over my own computer's software.

We've even managed to slow down things like Excel 97. I played with a 486 laptop recently, and was completely amazed at how fast things like Word felt (once they'd loaded off the ancient hard disk).

The most immediately notable part was the typing, which felt completely instantaneous. I haven't found out exactly why, but the switch from serial to USB keyboard interfaces (and the lag therein) as well as the various windowing pipeline "improvements" over the years seem to be in the mix.

on PCs, keyboards were never "serial" as in "the serial port"(mice were), but they were always "serial" in the same way that the USB port is "serial". Perhaps it's better said that the switch from a special-purpose interface to USB is what introduced lag.

I know on microcontrollers, if you want to reduce jitter (by sometimes up to a magnitude), you disable USB. Ive noticed my interrupt loops on ARMs go way down when the micro doesn't have to service USB.

But then you have to use an STLink to upload programs.

Old PS/2 keyboards are faster than USB because they were based on interrupts. That said, I think the cause underlying perceived low speed on modern platforms is more a bloat and inefficiency problem. Things are totally out of control. The PS/2 slight advantage over USB is an anecdote.

The Word processor on my 8 Bit Computer (BBC Micro) is ready to type into about 2 seconds from when I switch the power on.

Relevant post on keyboard latency here: https://danluu.com/keyboard-latency/

Takeaway: today’s keyboards are slow. Even gaming keyboards marketed as fast are slow. There’s more latency just in today’s keyboards than the entire end-to-end latency in an Apple II system from 1977.

There were recently some benchmarks of USB keyboards and they all add quite a lot of latency, even the so called gaming keyboards.

I feel like this is the core of how "Open Software" is aiming to combat this. By removing the consumer-driven model away from software and putting users in closer touch with developers, we are supposedly developing tools together that better fit our needs. Though, OSS arguably is struggling to meet this theoretical ideal, especially with the current lack of support most OSS devs have.

My life improved considerably when I switched from Sublime to vim, Excel to VisiData, Word to LaTeX (though, I am unsure if LaTeX is completely OSS), etc. I invested the time in learning the language of the tool, but it pays off in droves.

Sorry to say, but the big name DEs are off chasing the UX tail of the big name commercial stuff, come hell or high performance costs...

There's no difference between OSS and proprietary software for the users, except price and the (most often theoretical) ability to modify the software and redistribute it.

Don't get me wrong: I use OSS wherever possible, contribute where I can, and love the freedom to tinker with the software's innards. But UI just isn't OSS' strong suit. Mainstream Linux desktops are no faster than MacOS, and quality as perceived by end users is far lower.

Some of that is due to missing hardware support, and some of it is a result of the OSS' "bazaar model" of development simply not being as good a fit for UIs as it is for CLI applications. But in any case, the forces at play in the market are the same for OSS and proprietary software.

> I am unsure if LaTeX is completely OSS

The TeX sources are widely available and (infamously) well documented. The LaTeX macros are merely widely available. I forget what licence they're under, though.

What makes you unsure they're open source?

> This could be a copy-paste of rants I have written. I think marketing has ruined pretty much everything. Everything has a subscription now, and it's in the cloud.

Well, shrink-wrapped software only needed to be purchased once to have something, and it was amenable to piracy. It also didn't let companies collect, expropriate, and sell off your private data. Hell, once upon a time, people even wanted to take EULA provisions to court as illegitimate, since you couldn't use a product you'd already paid for if you didn't "agree" to a post-purchase "contract".

Something had to be done to make profits sustainable!

I would blame Scrum and the culture of storyboards. Go and take a look at any storyboard. 99% of it is feature addition. Not speed improvement, not security. It is some "scrum master" who comes up and says 10% of our customers have said that feature X on our app would be awesome and based on this POOMA-ed projection we will have a 10% improvement in retention and 100% improvement in some other random statistic, completely ignoring the fact that you are negative impacting things like speed and responsiveness that the other 90% of the user base gives a crap about, which was not captured in this random survey of yours

> I think marketing has ruined pretty much everything.

Actually, in this case, it was technological advancement coupled with those companies' desire to make profits that 'ruined pretty much everything'. Marketing had nothing to do with it.

In fact, do us a favour and re-read your post, then tell us what parts were ruined by marketing?

The cloud is DRM. I think that and not scalability or anything else is the primary driver for most apps. It's a way of putting your code where users can't directly access it and charging a monthly fee.

Atom is pretty terrible, but you have options. VSCode is also based on the web stack and is very fast (though Sublime is even faster).

Ugh. Guys think bigger than just web stacks. Pluma/Gedit, notepad++, TextMate... stop sucking off JS

Kate, Sublime, Vim, Emacs, 4coder, UltraEdit, BBEdit, Bluefish...

After using various IDEs for a brief period of time I was always comming back to Kate because there's no input lag there.

Sounds like you need a new editor, friend. Can I suggest Sublime Text?