Lol ok. You're right I don't know **** about what video encoding is or where it's done. But if you're arguing that iPhones are far superior for those apps, ignoring the awful Android porting that's done on them, you've lost all credibility here.

I don't see how app revenue has the slightest relevance to this either.

You're just making garbage tangential points to stir **** up. You're not actually arguing anything cohesive.

The "glorified" part is pretty important. There's a lot more things that need to get done by the "text editor" for programming. Good IDEs have a lot of features that involve looking at the whole code base at once. For example, when a change in one portion of the code breaks some other code and the IDE needs to notify the programmer. I don't think there's a lot of equivalent features in a non-programming text editor.

This also misses that programmers generally have to run their code. VMs, simulators, etc.

I have a modern macbook pro with and I still hit performance issues at times.


In terms of a lot of the rest of the discussion, it seems ridiculously simplified and based on a few weird premises. For example:

Even as a simplification this is just weird. Its not the lack of choice that make people focus on the negative. It's the lack of choice that makes the company not care about a lot of these 'minor' negatives.

Like, the reason Excel doesn't have an amazing auto-save is almost certainly not because they're worried about a 20 year old ecosystem (there's lots of ways to add the feature w/o ruining everything). Instead its probably just because they don't need it and its not worth their time and money to build the feature. It's not a particularly impossible feature (you can imagine Excel as a DB and just follow the models that DBs use to get really good auto-save functionality even in cases like Dave's crazy macro) but its not going to change much.

It starts next week, so I don't know yet! I saw you mention using it recently, any particular resources you recommend for getting started?

I assume in San Francisco at least that it's related to the size of the tech industry here as well as living costs. When I went to Stockholm earlier this year and found it to be so expensive I wondered if their tech salaries would be similar...guess not

Yeah but free healthcare, free college, 6 weeks off, pension for life, etc. No student loan burden makes a huge difference.

Not really. I just messed around and followed a guide to create a restful crud api. The language spec is really simple and short, so people advocate reading the complete thing. Not something you'd do for other languages.
I just did a comparison on my city (Reykjavik) and SF and found that most things were more expensive here (food, travel, entertainment, booze), but that rent was about double in SF.

Plus there are intangibles that are not represented in our salary like free-ish healthcare, cheap childcare, more vacation, more sick leave, parental leave, better work-life balance, etc.

SF rent is a ridiculous outlier.

But non-dominant products also have all sorts of these minor negatives. Even more of them, for sure, whatever keeps them from having 100% marketshare other than price is absolutely at least a minor negative. There's no endorsement stronger than someone continuing to use the product. And most of these minor negatives aren't even negatives but preferences. You cannot satisfy everyone. One mechanism I'm clearly not explaining very well is that non-dominant products tend to lose those users who are clearly bothered by these specific negatives. But in a similar situation, dominant products, due to their other superior features, are still stuck with users, who then talk about how terrible the products are. Despite still using them. This breeds resentment against the brand, even though clearly they are getting something out of using the products - no one is forced to use anything. So those other features that are compelling these users to use them, which makes the products objectively better, actually lead to more negativity against the products because in the absence of those killer features, those users would not even be using them and would have no reason to harbor and spread this negativity.

Excel has a great auto-save - it just doesn't have the exact auto-save that suzzer wants, which I don't think most people want over the current setup.

This is true of every product ever.

This doesn't help suzzer. His issue isn't that Excel doesn't have some fancy time travel auto-save - his issue is that the feature as it exists, which technically satisfies his needs, is too fancy and offers him a choice he doesn't want. He wants Excel to just overwrite everything automatically so that he doesn't have the choice of multiple versions. What Excel actually does is closer to what you're describing. The exact semantics you're implying are probably impossible, for the same reason that web browsers can't perfectly recover the exact state when they crash, but Excel is not far from it.

Also:

https://en.wikipedia.org/wiki/Choice-supportive_bias

And:

Just push Ctrl-S for me once a minute. Not that hard.

Ok, I think I don't understand how the feature itself works now then and/or what Suzzer wants.


Right, but its why you can't make claims like:


I don't believe this is true, but maybe just a misunderstanding on what semantics I'm implying or that you think I'm implying. Not a big deal though.

The exact thing suzzer wants absolutely breaks third-party excel integrations for the reason I already gave. Let's go through this in more detail. Let's say I developed an Excel Add-in that, on save, for some class of workbooks that are clearly designated as such, scans through the workbook and encrypts certain pieces of information that are clearly designated as such before saving the workbook so that no one without proper credentials or the add-in, when they open the workbook, could see this information. And let's say I used a beforeSave handler to accomplish this.

And let's say another developer developed an Excel Add-in that, on save, for some class of workbooks that are clearly designated as such, sends certain pieces of information that are clearly designated as such, to a web service, and depending on the response, modifies the workbook and halt the save process.

Well, you cannot satisfy both these add-ins here - you absolutely need to trigger beforeSave for autosaves in the first case and you should not trigger beforeSave in the second case. These are kind of terrible examples I came up with, but it literally took me one second after I saw what suzzer wanted to think of this and I'm sure someone who actually works on Excel may have a whole bunch of other concerns.

Then as a user don't use that auto-save option with fancy 3rd party stuff, or have the plugin warn the user or disable it or something. You're making this way more complicated than it needs to be.

You think people like me who Ctrl-S constantly and reflexively wouldn't run into the exact same issue? Users don't ever try to save when something is happening in the background? How does the plugin deal with it then? My generation grew up w/o any auto-save at all and most of us lost at least one multi-page school paper to the problem. So we compulsively Ctrl-S now. Just imagine the feature I want as one of those little tumbler bird things, except it can magically hit Ctrl, then S every time.

Almost everything you would want in regards to autosave pretty much exists:

https://www.ablebits.com/office-addi...ed-excel-file/

But suzzer is saying this is too complex and wants Excel to just overwrite (as though he clicked save) the current file every so often so that if Excel crashes, you can just open the file you were working on and it has the changes you need.

I would like to plug in an iPhone, open it's file system, and drop music onto the phone.

I would like to use a decent pair of headphones, like a Sennheiser (which cost considerably less than low-quality airpods) when I'm charging my phone.

Oh right, I can't.

And sometimes you hit ctr-s and lose an hour of work. That's the life of Excel. We took it for granted that we would lose 1/2 day of work every week.

If you constantly and reflexively Ctrl-S, why would you need Excel to do it for you?

This has nothing to do with something happening in the background - that was what daveT was talking about and it's a non-issue. It has to do with application semantics. People have already written applications assuming certain semantics around what it means to save - changing the semantics breaks the applications. Breaking APIs at this level can be okay if you're adding something substantial, but in this case, what you're proposing is just a less feature-rich version of what already exists. The current feature, as designed, already accomplishes everything you need, while also protecting people who don't want things overwritten, and while maintaining better compatibility with third-party add-ins. It also handles edge cases better (new workbooks that have never been saved, unreliable or slow network/removable drives, etc).

Like sublime does. I just want the option. You know it's technically feasible, you just think it's a dumb feature that no one should want. But still it would be about the 49,344th dumbest feature in the office suite.

Because I forget sometimes.



You totally skirted my question of why auto-hitting Ctrl-S during [XXX thing where it would be bad] is any different than a user hitting Ctrl-S during the same thing. Especially if auto-save is not default and I explicitly turn it on as a user, maybe even with a big warning that it might cause problems with a 50 million line macro or w/e.

Actually the other reason why they can't simply hit Ctrl-S is that one of the most common reasons for Excel crashes is faulty add-ins. So if there's a bug in some third-party add-in's beforeSave routine that causes it to crash, calling beforeSave to protect against crashes is kind of contradictory - not only would it crash before you save, it would massively increase the frequency of your crashes.

Well if you can't save you're kinda ****ed anyway right?

I don't think you realize what real Excel work entails. It's not a dumb feature, per se, to have auto-save, but it's a precarious thing to add into the system, at best.

And candybar brings up a good point about the plug-in ecosystem. For example, we had a plugin that remotely attached to some .Net system. We could pull down 150,000 rows, which, for argument's sake, took one hour. This is 60 ctr-s points, all while Excel is yanking data from outer space. When exactly should these save points run? Consider that there was absolutely zero reason to ever save a single row of the data we pulled. We made changes, pushed them back up, closed Excel, and rebooted the computer.

You're missing the point. I said that Apple isn't underperforming for the purpose of higher competition.

Regardless, I also believe that in Apple's vision, the file system restriction and lack of headphone jack actually yields a BETTER product. The logic is the improved security from the closed file system and slimmer hardware design/better battery life/whatever from the lack of headphone jack outweighs the negatives.

XXX thing where it would be bad is daveT's thing - has nothing to do with my point. My point is that Excel, on user-initiated save does this:

call any beforeSave event handlers
Save
call any afterSave event handlers.

And all kinds of add-ins attach all kinds of handlers to these events. And without knowing their intentions, it's impossible to discern whether triggering these handlers is appropriate or inappropriate. I already gave the examples. If you don't want the existing file to be overwritten without specific automatic things happening before, you need to trigger the handlers. If the handlers aren't about making sure that the file is in a good state before saving, but in fact are part of some sort of workflow in an interactive application that triggers an elaborate sequence of actions that the user expects from clicking save, then you absolutely don't want to invoke the handlers. And assuming that every Excel user is going to do ctrl-s reflexively is silly - some Excel users aren't even Excel users but are just using an enterprise application that happens to use Excel as a platform.

Maybe. I doubt its really as big of a deal as you say, and its also not unheard of for companies to break existing integrations from one release to another.

It's all irrelevant to my point though. You made a claim of fact something that is far from factual. We have no idea why Microsoft didn't build the feature - and its much more likely it was a simple reason (like we agreed applies to like every product ever) then your much more specific assertion.

I have to share my music with people, and I always know who is using an iPhone.

That's the thing I don't understand about the iPhone. It's presented as some sort of music / cool / hipster / lifestyle thing, but you can't easily do anything that involves anything related to the things they advertise.

Sure, protect for security, but too much security sacrifices "easy to use" and "just works."

Here's a music professional complaining about the headphone jack here:

https://news.ycombinator.com/item?id=15402091

In fact, many music professionals are complaining about all the things Apple has been doing lately. They were the people who stuck through the thin years and are the ones who really were Apple's loudest proponents. See YouTube for various rants.

One of my friends is a video editor. He's been using nothing but Apple and Apple products since the 80s, and he never used anything else. He bought an Android earlier this year and is totally smitten by it and will never go back. He's also considering moving over to Windows for his next computer. Apple products are expensive, and the only people who really needed this stuff (until fairly recently) was professionals.

Hell, I need to get a new computer for music recording, etc. I was considering an Apple, but now I'm having a lot of doubts about it, and Apple users are telling me to not bother with it anymore.