partial_accumen

I’ve worked hard to gain salary only to find that it didn’t matter. Every review I’ve ever had was a lie. If I was given a good raise, I was told that it was my hard work. If it was a bad raise, they found one item to give me ‘satisfactory’. A bunch of us shared our salaries over drinks one evening and we all were about the same. That was a big surprise to me.

You’re proving my original point. Staying at a single employer for years, and you’ll get minuscule raises irrespective of the level of your efforts. Further, you get a sham of job security. When tough times come (as I’ve seen three cycles in IT), layoffs can come and take your job anyway. Without having built up a war chest of savings to live on, your living situation and that of your family is at risk.

Back to the point of the original article, employees talking is bad for employers. Unionization is one way to solve the collective agreement problem, but there are others. When employees (or any group for that matter) organize, they can make things happen.

You’re trying to fix employers. You’re welcome to go that route. In the original article Doctorow posited that “vocational awe” was the reason IT people put up with such conditions. Apparently that’s true for some. However, I also know it was not true of others who preferred to make the money they needed to eventually stop working for someone else.

I had a 13 year old account I simply stopped using in 2023 when compact.reddit.com was killed and the API changes fees were introduced. Lemmy was “good enough” get to get what I was getting from reddit. I stopped using reddit and haven’t posted since.

It feels like fear mongering when there are no data to back it up (this is not a knock against your post, it’s a complaint against the argument against unionization).

If an org, under union influence, would do 15% to 400% salary increases year over year for their entire company/department, they’d likely go bankrupt. Yet that was possible on an individual level without a union in place. I didn’t really mention it before but employers that treated their workers poorly were many times an asset to this method. That bad behavior drove away workers, meaning the bad employers would have to increase their salary offerings much higher to attract a worker to join even with the bad behaving employer. It also meant that the IT worker, who may not have been entirely qualified, would would have a shot at getting the position (and become qualified on the job). Once that that worker is qualified (after the year or two), they can take that knowledge and experience and jump ship to a good quality employer, gaining yet another with a big raise. The worker also just collapsed 5 to 10 years of slow career growth into 1 or 2 years.

I only know one person in a union and they have limited anecdotal data that shows that the cost of being in a union is offset by salary gains.

I’m guessing those quotes are about salary gains across a the entire company/department. This was nearly mercenary-mindset IT work. As in:

• Get in with the raise
• Learn the next thing you need
• Work the thing for a bit until you know it and have the experience and expertise at that employer
• Get out

Rinse repeat.

None of that is assisted with a collectivist union mindset or union implemented rules. Please correct me if I get any of the following union benefit bullet points wrong. As I understand it, the union would do everything to undo that situation. They’d:

• work to normalize pay across workers fairly.
• emphasize a proper work/life balance
• enforce conflict resolution with strong worker advocacy
• encourage/provide training across the company/department for continued competency among everyone
• establish rigid rules for promotion

IT has been a raging river, but if you were able to navigate it, you’d get to the end very quickly. You’d certainly come out with some cuts and bruises though. If getting to the end (comfortable money in our case) is what you were looking for, then it was the fastest way to it.

Shhh! We’re releasing that accuracy update in the next version of the product. We need to sell through our existing inventory of the less accurate ones.

I have a competing technology that is nearly as accurate. For only $50 I’ll send you this device that you will have unlimited license usage rights to. While not 53% accurate like my competitor, its proven by scientific studies to be 50% accurate. I also offer volume discounts if you buy 10 the price drops to only $45 per device. Sign up now!

Its not in place yet. I’m seeing it take shape. I don’t see how it can be successful on its own and will either lead to user teams adding their own Shadow IT or the skeleton IT Operations group balloon into a giant Shared Services outfit.

However, I thought I’d mention it as its the same mindset that lead to Devops, which was just a business reaction to finding a way to hire and maintain fewer IT roles. Its interesting to see the IT roles being pushed directly onto the users now.

If your users are external to the org then this probably doesn’t apply. However, if your users are internal, you give them a repo of their own and grant them access to publish to the pipeline.

Devops for sure. (“Why have IT people when we can just make developers do it?” Fucking brilliant ☹️)

I’m not sure if you’re tracking Enterprise IT trends these days, but its evolving yet again with “Why have Devops people when we can just make the USERS do it?”

Suffice to say, those of us that know how to clean up messes (or realistically become Shadow IT) will have gainful employment for the foreseeable future.

Pease tell me you know of someone where this actually was true:

I know many personally that earned this kind of money.

that they made crazy money and they’re set for life.

Yes, but only for some of those because most chose lifestyle increases instead of savings. So I know tens of the folks that can retire right now and have more than twice the median income of the USA for their entire life without running out of money. This means they’d have a couple million in the bank. This would be an annual retirement income of about $80k-$90k/year USD for one person (thats also without any Social Security benefit added on top yet). Many times a married couple both work in tech so that would be household retirement income of $160k-$180k/year for as long as they live (again I’m not even adding in Social Security benefits in this yet) if they chose to retire early. These folks have another 7 to 15 years in the workforce if they choose to retire in their mid 60s.

Thats not “chartering jets to Monaco” money, but its a very comfortable retirement. That was realistically attainable for lots and lots of people in tech after working for 15-25 years.

Because, based on 30 years in and a complete lack of knowledge of anyone who got out and retired early, either personally and via someone I know,

There were lots of IT jobs that paid decently (not amazingly) and if you worked for that one company for 20 years you would NOT have the money I’m talking about. To get this in IT required chasing newer technologies (or sometimes chasing specific older ones), and changing jobs frequently, usually every year or two but sometimes as short as only a few months. The trick is (which took me longer to figure out than I’d like to admit) that corporate annual raises were minuscule. You’d work your ass off for maybe a 2%-5% raise each year. Whereas if you did the exact same work and effort and changed jobs your new employer would give you essentially as low as 15% and sometimes as high as 400% salary increases in a year. You can quickly imagine that you only have to do this a handful of times for your annual salary to be huge. Think about how long it would take you to have a million dollars if you were making $180k to $400k and invest that savings ten years ago.

I know many one that has actually chosen to retire early even though many have the funds to do so. When you’re at the end, you’re at your peak earning phase. So working just a few more years means massive increases in retirement income. This is the bit of a trap that keeps you working voluntarily. Sometimes retirement gets forced early with RIF/redundancy/layoffs in your 60s and you may not be able to get another tech job however.

The even more risky path if you want the many multimillionaire path, is you started your own business. High risk of failure, especially tech types that don’t know the business side, but for those that have the right partners/help and can thread the needle, this is the tech path to the tens of millions of dollars of wealth.

I conclude the only people for whom this worked were C-level. Even the smartest man I know didn’t cash in and get out.

I’m talking all non-C level here, just technologists/individual contributors. At most they might be project or tech leads, but they’re still not executives. I don’t know the executive C-level advancement track so I can’t speak to it. Maybe the folks you know were the ones Doctorow was talking about. If the smartest man you know didn’t know to trade up or cash out, then maybe he was one that was in it for “vocational awe”. Do you have any knowledge as to why he didn’t trade up or cash out?

All respect to Mr Doctorow, but he’s got this wrong:

Tech workers are workers, and they once held the line against enshittification, refusing to break the things they’d built for their bosses in meaningless all-nighters motivated by vocational awe.

…and…

Tech workers stayed at the office for every hour that god sent, skipping their parents’ funerals and their kids’ graduations to ship on time.

It wasn’t “vocational awe” it was money that lead tech workers to work long hours and sacrifice. Lots and lots of money, five to ten times what your non-tech same-aged peers were getting. It was so much money that if you didn’t live too high on the hog, it set you up for a very nice retirement and having “fuck you” money in your late 30s and 40s. During those days the only thing a tech union would do would make your life balance better, but at the cost of your salary.

With all the tech layoffs and enshitification, those meteoric salaries are starting to come down to Earth. They’re still high comparatively to other professions though. So I think tech unions will gain more traction now, but employers also have more tech workers (right now) so they can bully their current workers to try to avoid unions. However, tech is cyclical, as is hiring. I’ve been in tech long enough to see 3 large downturns, but when the pendulum swings, the hiring returns and (so far) those high salaries have too. If the pendulum swings too quickly and the high salaries (and now “work from home” requirement) returns, tech unions will be back to where they were struggling to establish themselves in the industry of job hoppers jumping ship from one employer in under a year or less chasing the larger compensation.

We can do this without anyone dying, as long as we all work together.

Have you met humanity? We don’t do “all work together”.

Lots of people will likely die if this goes forward. You’re okay with that? Are you okay with being one of the dead?

Its not rolling back the clock imo. We have already pulled out those resources. They are in our possession. We don’t need to mine fresh rare earth metals.

So if your policy goes into place, all extraction of rare earth materials stops at, lets say, midnight. We’ve got some spares on the shelf, but without replenishment, and knowing that replen will never come from virgin materials again, those components are horded.

With the global knowledge of this, industry and consumers rush to buy up remaining stock. In three months most electronics stores will be bare electronics. This includes mobile phone stores too. In about 2 months we’ll see new automobile supplies dry up because specific critical control modules simply can’t be built new anymore. Most cars on the road today will break, and simply be parked or scrapped because replacement parts are simply non-existent.

Existing deployed systems all over the world will start to break and not be fixed anymore. Simple things like digital signage at stores will break and remain dead or be ripped out altogether. Lines (queues) will be much longer as many kiosk driven activities now have to be done by humans. Think airport or train check in. Delays in post or package shipping will increase as transport infrastructure starts to break down.

Most of the western world will still have food for many months, but variety will decline dramatically. Anything delivered by aircraft will suddenly cost much MUCH more money because carriers will be trying to keep low hours on now (mostly) un-repairable aircraft.
New computers will start getting bigger again and slower again. Much of the benefits of these materials making computers smaller, faster, and require less electricity.

It will take probably a decade for your recycling program to come online at any scale that can replace what we have for supply chain right now. Even then, recycling can’t easily replace some of the materials as they are bonded chemically during time of manufacture so many lower cost semiconductors simply stop being made.

None of this speaks to the massive economic impact to the world where tens of millions of jobs start to disappear because the world they did relied on affordable devices which are now a premium priced item. Economic upheaval felt by this will make the tariff war we’re going through right now seem like an ideal fantasy.

It will be very eerie to watch our societies and technologies slowly crumble before our eyes and things that were considered near throwaways be now treasured relicts of the past of an age of abundance.

Thats the thing. Compute ENABLES all of that industry. Compute ENABLES all that residential energy usage. Compute ENABLES commercial businesses to operate.

Those sources are referring to direct energy usage, but isn’t accounting for indirect production enabled by that technology.

I agree with this, but this is why I asked you the second question in my original post to you. That question was “And all the services you consume that use these?”

So if you’re saying “yes” to that too, then you’re essentially wanting to roll the clock back to life back in 1940 or so. The consequences on human life will be devastating if we do that. It may be cutting the human population on Earth in half. A good chunk of what compute enables is human life.

I am not going to lie, i just don’t have the energy to put together all the research that has been done on the energy consumption of AI neural networks for you.

I did all that before i answered you because I wanted to make sure my thinking was accurate.

Total electricity (not energy, because energy is oil, gas, coal, etc, tool, just talking electricity here) used by all data centers in the USA for all computing is about 5% of the total USA electricity consumption. source.

They are consuming more energy than entire States in the USA use.

True, but I’m not sure what relevance that has. Lots and LOTS of industries use way WAY more electricity:

“The industrial sector accounts for 33% of all the electricity used in the world (the largest sector for energy consumption is residential housing, followed by commercial businesses). According to the U.S Energy Information Administration, in the United States, 77% of all industrial electricity goes to manufacturing, 12% to mining, 7% to construction, and 5% to agriculture. From the list of high energy consumption industries in manufacturing, chemicals account for 37%, followed by petroleum and coal products at 22%, paper and paper products at 11%, primary metals 8% and the remainder is made up of food, non-metallic metals and all other categories.”

source

Then theres cryptocurrency.

I’ll agree its a waste, but of the 4.4% used by ALL datacenters 1.5% of that is crypto.

AI, by itself, last year was about 1% of the 4.4%.

If you’re rationally focused on CO2 reduction, all of compute is a drop in the bucket compared to a number of heavy industries. Again, my numbers here was only about electricity, which does have CO2 concerns, but lots of those other industries use way more electricity in addition to other CO2 producing energy sources (like natural gas/coal/oil, etc).

So if your true goal and concern is CO2 reduction, your priorities should be going after the much bigger fish.

You don’t think nations in Africa or Asia use computers, electronics, telecommunications equipment, medical device for things like imaging or chemical analysis in their logistics or supply chains?

Ahh! I understand now! You didn’t read the thread you’re responding to where the OP said these metals shouldn’t be mined at all. You just jumped in and provided an answer for a question you didn’t understand, then you attack my response because of YOUR misunderstanding. You think you’re responding to a tariff question, and not the OPs position of climate change.

Please try to read what you’re replying to next time before you make yourself look foolish like this again.

A good improvement, but an important distinction from the headline alone:

This energy savings comes with a caveat. “It is sort of a best case because the 30 percent applies to the network stack or communication part of it,” Karsten explains. “If an application primarily does that, then it will see 30 percent improvement. If the application does a lot of other things and only occasionally uses the network, then the 30 percent will shrink to a smaller value.”

Otherwise, ya, hermit life FTW

It won’t be hermitage for us, it will likely be death from starvation and disease. Global supply chains, including those for food production and distribution required modern technology. If you’re going back to pre-computer world you have to roll the clock back for how much of a population the world was able to support. The first transistor was made in 1947, which is arguably the beginning of modern electronics (a few vacuum tube computers existed before this time).

World population a few years later in 1951 was 2,536,927,035. The world population today is 8,231,613,070. So your suggested change will kill off about 5,694,686,035. Even Marvel’s Thanos was only trying to kill off half of the population, and here you are suggesting Thanos wasn’t going far enough where you want about 70% of everyone dead.

To think either one of use would survive is hubris.

What you’re suggesting is “pennywise and pound foolish” if your goal is CO2 reduction. Semiconductors and Rare Earth metals are required for our best weapons against CO2 emissions, those being wind turbines and PV solar panels.

We are eroding our planet of its life by causing drastic changes to our climate, caused by burning up massive amounts of fossil fuels to power our “AI”.

Energy demands are far FAR larger than the minuscule (by comparison to other energy users) AI data center waste.

Not stopping now will only add more Co2 emissions to our planet’s atmosphere, which will increase global temperatures.

Fossil fuel lovers will back you 100% on reduction of semiconductors as it means a lock-in for electricity generation to mostly fossil fuels.

So no modern electronics or computers then? So back to maybe basic transistors at best? You’re cool with that?