Re: How is IBM still a thing?
Its easier to think of IBM as a 20 different companies, rather than as an entity. Some of them are good to deal with. Some, not so much.
21 publicly visible posts • joined 4 May 2022
Is that going to make my PC substantially cheaper? Probably not. It's a full-stop NO for anyone who games. You need x86 on Intel or AMD and either NVIDIA or AMD for graphics. Game producers are not going to make ARM for Windows versions of their AAA titles when the majority of windows PCs are x86.....and ARM needs top-end gaming to become the majority of Windows PCs. We've seen this before in history.
Commodore's 64 was incredibly popular. So they made the Commodore 128 - faster chip, more memory, more features....and a C64 mode for C64 software. Software vendors looked and said "well I can write code for the C64 and have the entire Commodore market available to me....or I can write for the C128 and get a small fraction. Guess which one they wrote for?
ARM is certainly not going to make it easier to work on or more compatible. It will be worse.
In enterprise environments - they might have a shot with cheaper - but there will always be house-written software or old software, or something that means headaches to switch to ARM. PCs for corporate desktops are pretty cheap anyway. I don't think anyone is going to trade compatibility headaches for a twenty percent discount on the hardware alone.
Uhh....no. There are millions of dollars a year for 24/7 365 support, ongoing license subscriptions, and consulting.
Buying core networking equipment, which is critical infrastructure, isn't like buying a TV at Wal-Mart.
Plus, a ban by the government will lead to a ban for *suppliers* to the government, for the same reason. So existing and future customers lost. Total inability to compete in the German market.
Ericsson, Nokia, Cisco, Juniper, HPE......all of them stand to gain. ZTE and Huawei loses.
Having an 'elected official' isn't a good thing.
They are subject to pressure. No, not the pressure from the little guy, pressure from business, donors, etc, because of the stupid way campaign finance laws are set up in the U.S.
On top of that 'elected officials' rarely have any experience or education that would make them suitable for the job. i.e. if it wasn't an elected position, that person wouldn't have a chance in hell of getting that job. Think Majorie Taylor Greene.
Sigh.
You don't need to restrict the *employee* to fix that.
Any good consulting contract has a poaching fee to keep companies from doing just that, or even an outright ban on the customer stealing the employee. Its pretty standard contract language. You don't need to restrict a person's right to make a living to achieve that goal.
Remote desktops have been around for years, and there are a few use cases for them. But its never worked out - CPU and memory on a server is a hell of a lot more expensive than it is on a desktop or laptop. The numbers *never* work out in most cases. Plus, they always suck. Always.
The easiest way to get excited beancounters and pinhead executives to backtrack away from this is to ensure that THEY are the first users. Won't take a week before they opt out.
Any car that does that would be immediately off the buy list.
Any car that requires subscriptions to use common features....immediately off the buy list.
Next up: Cars that automatically report driving info (hard stops, acceleration, distances, etc) to insurance companies. Fuck them too.
From a software compatibility aspect and from a support aspect, the move to to a linux desktop is more trouble than its worth. Yes, LibreOffice is a great product. Except nobody outside the company uses it, so suddenly we have to deal with formatting issues between MS and Libre. And that doesn't even cover specialty software used for various hardware or business processes.
Is Linux probably more secure? Yes, likely. Is it worth the headache of linux on the desktop for a large IT department? No. Not even close.
Linux on the desktop is a dead-end road for enterprise desktops, no matter how much everyone collectively hates Microsoft.
The other big problem is actually the commodity hardware. Part of the sales pitch was being able to turn on services as needed....which meant over-provisioning. The reality was that customers rarely turned on new services. That meant a lot of horsepower sitting unused in the field, raising operator costs with no real way to recoup.
On top of that, many of the more proprietary CPE are...actually not that expensive. Add on your excellent points about single-pane-of-glass management and cloudification.....well NFV is essentially dead.