Though you'll need to deduct a lot of SG&A, etc. overheads from that as well, and it often looks far worse.
Big Tech companies (ie ones close to monopolies) still extract a ton of net cash flow that ends up going to investors and top management, though you could then ask why those monopolies extract cash from their customers as well, etc.
My worry with tech unionization is that generally it slows productivity increases and change. I get why bus drivers, etc. in stable systems should organize and could do so without adversely affecting system performance, but in tech/startups, I don't think those companies would exist in unionized form for very long before being put out of business.
Now they're just saddled with too much bureaucracy and politics, and that's already led most of them to underperform, at least on an innovation basis.
> My worry with tech unionization is that generally it slows productivity increases and change.
Do you have examples?
> Now they're just saddled with too much bureaucracy and politics, and that's already led most of them to underperform, at least on an innovation basis.
Yea, it's more complicated than that, but in the US context unionization (like in the automotive and steel industries) was shown to slow innovation. [1]
The counter to this is that Europe actually has better outcomes because it takes a more collaborative approach vs. the adversarial approach in the US, and this better approach has shown to improve some outcomes.
The linked paper is strongly biased towards the ownership class--and the think tank is labeled "libertarian-conservative" on wikipedia, so I'm skeptical of the conclusions.
Yea, almost anyone in the US asking about the productivity implications of unionization will be coming from the ownership class, as US unions don't see that as their problem vs. building worker power.
I couldn't find a paper about US unions increasing productivity. Can you?
Fair. And I didn't look for other papers. If the discussion is simply "unions decrease productivity" however that's defined, that may be true. But by how much? Is that a problem? To whom?
I suspect unionization is better for more people overall.
This can be true, but also many of these companies are themselves essentially monopolies that also don't face market pressure - such as Visa, Comcast, etc.
You can say this is due to corporatism, which is often true as well, but gov't regulation is required for natural monopolies.
This is to avoid issues like the early railroad boom where they hugely overbuilt, a lot of them went under and/or eventually coalesced into a few monopolies, which needed regulation to prevent them squeezing their customers to death.
The best markets are sensibly regulated. It's a balance, not a binary.
I'm always trying to figure out why OpenSCAD has such a following when, at least to my manufacturer-brain says "whoa it's inconvenient to use to make real stuff" and "there are more expressive 3D modeling DSLs".
Is this like hobby only like writing an app in brainfuck? Or am I missing something?
I think it’s the super low barrier to entry. When I first started 3D printing, I wanted to make basic shapes with decent dimensional accuracy. OpenSCAD lets you install a single app and start doing that immediately. It was easier for me to quickly learn the OpenSCAD syntax than it was to spin up on FreeCAD or Onshape.
I started with Tinkercad, which is online and WYSIWYG. Nice for newcomers, but as a programmer I felt that I needed something that could be more automated.
If I was dealing with diagonals, I needed to calculate sines and cosines outside and then place the values. If needed several objects with the same shape I needed copy and paste them, and if the base shape was wrong, I needed to delete them all, fix, and then copy and paste again. And if a value of dimension, position or angle depended on another it was more and more calculations. Soon it became unsustainable for me.
OpenSCAD saved me. Being able to create modules, using parametric values, declaring functions and so on made my life really easier. At the first it looked like it would be boring to do everything by code, but once you get used to it is actually faster than using a WYSIWYG tool.
There’s not many free as in open source alternatives out there. There’s traditional CAD such as Fusion 360 which gives you a limited amount of designs and is “free” for non-commercial use.
So then under the open source, own your designs umbrella you have stuff like FreeCAD which is similar to traditional CAD which means you have a learning curve.
OpenSCAD is programmatic which suits someone with a software engineering background. Plus being free in the sense of owning your designs.
> OpenSCAD is programmatic which suits someone with a software engineering background
Perfect. I already knew how to code, so it was just a question of minutes to learn the basic stuff of the language and then I could just do what I wanted.
Fusion360, AutoCAD and organic modeling tools like Blender you need to learn several concepts, keybindings, navigating through menus, panels, tabs, workspaces, etc, etc. So many things _before_ doing something. In OpenSCAD you can even code in the text editor you're already familiar and then use it only to render the object (which can also be done through CLI!)
Well, I personally just can't do 3D modelling any other way, if I had to use graphical tools to create it, it would be pain. On the other hand, writing code is a second nature for me, so even if it's a little inconvenient, it's still much better than anything else (for me).
I only used it once for modeling some procedural decorations in the shape of flowers. I did it on a long train ride and managed to both learn the language and do the models, so there's something to be said about the simplicity. If I had to do anything more complex or less artistic I would use other tools. Something python-based would be ideal.
OpenSCAD is not a programming language with flow of execution, but static statement of properties of the desired object. Only language better for this might be Prolog.
Once you totally comprehend this, all the "functional" shit of OpenSCAD makes total sense.
Python does not have that kind static description properties, and those OpenSCAD-Python translators suck very much, imho.
You still have to explain the magic bullet impossibilities, fully intact bullet on the stretcher, Oswald as known CIA asset (CIA reports show this), etc.
None of this is really explainable by the official story.
> You still have to explain the magic bullet impossibilities
This has been explained ad nauseam. The bullet went in a straight line.
> Oswald as known CIA asset (CIA reports show this)
I’m sure the CIA has used a lot of unstable people all over. It’s not inconceivable that one of them went on to commit an assassination without being directed to by the CIA. Sometimes things are just boring.
I completely agree that this is the most likely case.
That said, the public deserves to know the extent of the CIA's involvement.
I would be wholly unsurprised if it turned out to be some Mujahideen type deal where taxpayers invested a bunch to up-skill this guy, left him alone once the reason for the investment was over and he eventually came back around to shoot at us.
Sure, maybe the CIA did it, maybe the Cubans or Soviets did it. I'm not really invested in the "real truth" of the situation. I'm just saying that I don't think the remaining documents are going to say anything that we don't already know.
I'd wager that a 19yo is MORE likely than a 29yo, which is also relevant. Also, properly trained individuals are less likely to fall into this trap. How much training have these completely non-vetted children had?
