Everyone who knows that a technology where decentralization is one of the requirements will know this, and they will also know that this is a required tradeoff when it comes to decentralization - no free lunch.
A lot of people skewer blockchain-based tech for ineffeciency, but not a lot of those people realize it's supposed to be like that.
I'm glad that people are at least starting to realize that a lot of decentralized tech has the prerequisite of forfeiting efficiency.
And not just tech. In the article I tried to compare it to gardening.
But my initial idea was to compare it two political issues recently in my country. That story was far too unrecognizable though.
The issues are, where a decentralized police, that was governed by local mayors, was replaced by a central police force, governed by national government. Centralization that was sold as being much cheaper and more efficient. The cheaper was a lie, but only because the conversion itself turned out far more difficult. The more efficient is true. But only for a certain part of efficient.
Previously, e.g. mayors could micro target issues that were directly plagueing a small neighborhood. Now, this is still possible, but has to move through layers of bureaucracy first. In many parts, i'm told, it's less efficient. Overall, the sum, however obviously operates more efficient. Instead of each of the thousands of municipalities having some HR, there now is only one. Same for IT, etc etc.
The other issue is exactly reversed: mental service for Youth (youth care) a rather large institute, was decentralized: moved to municipalities. The promise was that this would allow far better detailed care. It is a fiasco, because the national government never considered that it would be far more expensive. They even promised it would cut costs.
Why voters fall for this, and think that both promise can be true (decentralized being more expensive and saving money) is beyond me. But the effect of decentralization on effincy is clearly visible here too.
I haven't observed any advantages to an office beyond overexposing the company to real estate (not really an advantage) - there is nothing I could do in person that I haven't been able to do faster in MS Teams (and better - now I have a written record and recording of everything). Another advantage of remote is that nobody ruins my train of thought by tapping me on the shoulder and asking stupid questions like "hey did you get that email." Best of all, no stupid pressure to go to lunch with anyone
I think the reality is some people do better remote, some better in office. There's no one size fits all solution. I hated my time working from home, but it sounds like you enjoy it. There's no perfect solution for companies here except to cater to individual preferences, or mandate one at the risk of angering some employees
There is one an individual might find works better for them. Another tractor is cost add employee market. I have a much bigger talent pool to choose from when hiring remote and I can recruite from much cheaper locations.
Of course teams work better in person. So many factors that impact that though.
Very little people are advocating that WFH doesn't work better for focus work of clear and specced tasks.
However, in my experience that is maybe 30-90% of software engineering, depending on environment, role and seniority. There are lots of relevant activities that are more challenging remotely, such as 1-on-1 mentoring, pairing for debugging or programming, or designing and collaborating a project or interface across-teams. Maybe you are "lucky" enough to never have to bother with that.
> 1-on-1 mentoring, pairing for debugging or programming
I find these quite easy or even better remote. I just fire up a screen share or even better a Visual Studio Live Share and both of us can see the same thing while we have our own developer environment.
Take this from someone who works effectively fully remote except for a couple days a month that I go into the office: Believe it or not, some people like their coworkers and want to eat lunch with them. Humans have been hanging out with each other since (checks math) the beginning.
I think brainstorming, discovery, and discussion type of tasks work much better in-person, maybe 5-10x better (driven in part by a small percentage that are thousands of times better).
Delivery works great remotely, probably better for me and many.
If you want to mimic the in–office fast paced decision making then yes in–person works better.
If you adopt a written RFC based slower decision making where people have time to fully articulate their thoughts and have time to respond to those comments then it not worst, just different.
If the house is burning and we need to come up with a solution now the first works way better. There is a reason why the military have war rooms. But for long term changes the second can be as good if not better.
I love WFH (and probably will not ever go back to the office), my team would all also say the same thing. Individually we will all tell you we are more productive. However, I know for a fact that my team as a whole is less productive now than pre-pandemic when we were 90%+ in office.
I also know other teams in our organization in the same boat. Sooner or later leadership is going to realize it and the gravy train will end.
The average Twitter user likely does not know what a file is (see: that one iPad commercial where the kid says "what's a file?" or something), and likely uses a mobile device of some kind (tablet or phone) as their daily driver.
Having unusual new concepts you need to explain is a negative point for sure, but that's a world away from the parent comment saying that having any new concept at all means it's doomed entirely.
Twitter had plenty of new concepts with replies, quote tweets, hashtags, threads, ratios. The evidence is that people can figure out this amount of new stuff.
Mastodon is a bunch of separate federated Mastodon instances. In order to see content from several you need to connect to each of them (if memory serves, your identity and social graph are not portable between instances due to the limitations of ActivityPub. This is a problem Dorsey's @ Protocol attempts to solve). That does let you see the illusion of centralization though, sort of like email which is also federated
The technical barrier to entry ensures "normies" will not be able to use or understand (or at least enjoy) Mastodon. This will prevent it from becoming as huge as Twitter
The more you measure productivity, the less productivity there will be. Employees spend a good amount of time documenting their productivity for nonsense like performance reviews, and that is a lot of time that could have been spent doing actual work
Classic counterintuitive anecdote where measurement increases productivity: https://en.m.wikipedia.org/wiki/Hawthorne_effect “The Hawthorne Works had commissioned a study to determine if its workers would become more productive in higher or lower levels of light. The workers' productivity seemed to improve when changes were made, and slumped when the study ended. It was suggested that the productivity gain occurred as a result of the motivational effect on the workers of the interest being shown in them.”.
Were you intoxicated when you wrote this, or am I just really tired? I'm having trouble connecting the dots on what you're saying. It seems all over the place
Honestly the fact that more tools are needed to make Mastodon usable shows that federation technology is "not there yet." I hope we are able to improve stuff to the point where it is usable.
Something I'm keeping an eye on is the @ protocol that is being designed specifically for the creation of federated social media applications -- it allows for portable identity and your social graph is portable, these things are not tethered to an instance of something. Hopefully that will be an upgrade so there will be less "jury-rigging" like this required to make federated applications usable
Everyone who knows that a technology where decentralization is one of the requirements will know this, and they will also know that this is a required tradeoff when it comes to decentralization - no free lunch.
A lot of people skewer blockchain-based tech for ineffeciency, but not a lot of those people realize it's supposed to be like that.
I'm glad that people are at least starting to realize that a lot of decentralized tech has the prerequisite of forfeiting efficiency.