Joey 
"[@reversalmushroom":](/forums/meta/topics/site-development-notification-and-feedback-thread?post_id=4650211#post_4650211
)
The last time we switched servers was in Spring of 2018, when we moved from the "Chrysalis" server to "Fleetfoot". If it takes you a year and a half to complain about it, then obviously it isn't that bad.
Though I assume you're actually referring to the site's software instead, since that was changed just a few days ago.
Byte's been working his ass off for the past 4-5 months rewriting the software from scratch. This isn't something he did for fun either - he didn't just wake up one day and was like "I'm bored, so I'm going to rewrite literally everything". This is something the site desperately needed.
Most people don't realize this, but Derpibooru is the largest site in the MLP fandom, and has held that rank for quite some time. We have around 1.2 million unique visitors a month to our primary domain, and see over 800 million requests per month between our primary domain and CDN domain.
And believe it or not - the site's been growing. It's very slow growth, but the show ending actually increased the number of people coming to the site. In fact, I've been seeing a ton of new artists just starting off with MLP artwork, or older artists who "retired" or "moved on" coming back and drawing pony art again.
All of this takes a toll on the single mid-range server that we run the site off of. We don't run this site for profit - we get enough through Patreon to cover our costs, but we can't afford some high-end multi-server cluster or an AWS/Azure infrastructure. We gotta make what we have work.
The old software worked well, but it was relatively slow by today's standards. It also was far from efficient on system resources. The purpose of the rewrite was to make the site load faster and reduce load on the server. It's been successful at both things.
If you wanna talk about a negative impact on user experience - just look at Derpibooru's predecessor, Ponibooru. That site was constantly overloaded to the point when it actually was up and working, pages would routinely take over thirty seconds to load. While we were nowhere near close that bad, slow load times and high load are a great way to ruin any site.
And yeah, there's been a few bumps along the way. We had an open beta, and while issues were reported and fixed, there are some things that just don't show until it's fully live. Replacing a system this big and complicated with minimum downtime is no small undertaking, especially for a single person.
Byte, quite frankly, has rocked this update. He's been working very hard to keep everything running smoothly and address issues as quickly as possible (to the point where he's making me"[concerned about his health":](https://i.imgur.com/N3GBzoJ.png) sometimes). Your negativity is not helping. If you spot an issue, "[report it":](/forums/meta/topics/report-site-bugs-here) and it'll be fixed in an order based on impact and priority.
If you just wanna complain for the sake of complaining, go vent somewhere else.
Finally, this update was primarily backend. There were a few frontend changes, such as for optimization or just some cleanup/to-do items that were easy to address during a rewrite, but the intent was not to have some kinda drastic change in the user experience beyond faster load times.
PM me your cute OCs
The last time we switched servers was in Spring of 2018, when we moved from the "Chrysalis" server to "Fleetfoot". If it takes you a year and a half to complain about it, then obviously it isn't that bad.
Though I assume you're actually referring to the site's software instead, since that was changed just a few days ago.
Byte's been working his ass off for the past 4-5 months rewriting the software from scratch. This isn't something he did for fun either - he didn't just wake up one day and was like "I'm bored, so I'm going to rewrite literally everything". This is something the site desperately needed.
Most people don't realize this, but Derpibooru is the largest site in the MLP fandom, and has held that rank for quite some time. We have around 1.2 million unique visitors a month to our primary domain, and see over 800 million requests per month between our primary domain and CDN domain.
And believe it or not - the site's been growing. It's very slow growth, but the show ending actually increased the number of people coming to the site. In fact, I've been seeing a ton of new artists just starting off with MLP artwork, or older artists who "retired" or "moved on" coming back and drawing pony art again.
All of this takes a toll on the single mid-range server that we run the site off of. We don't run this site for profit - we get enough through Patreon to cover our costs, but we can't afford some high-end multi-server cluster or an AWS/Azure infrastructure. We gotta make what we have work.
The old software worked well, but it was relatively slow by today's standards. It also was far from efficient on system resources. The purpose of the rewrite was to make the site load faster and reduce load on the server. It's been successful at both things.
If you wanna talk about a negative impact on user experience - just look at Derpibooru's predecessor, Ponibooru. That site was constantly overloaded to the point when it actually was up and working, pages would routinely take over thirty seconds to load. While we were nowhere near close that bad, slow load times and high load are a great way to ruin any site.
And yeah, there's been a few bumps along the way. We had an open beta, and while issues were reported and fixed, there are some things that just don't show until it's fully live. Replacing a system this big and complicated with minimum downtime is no small undertaking, especially for a single person.
Byte, quite frankly, has rocked this update. He's been working very hard to keep everything running smoothly and address issues as quickly as possible (to the point where he's making me
If you just wanna complain for the sake of complaining, go vent somewhere else.
Finally, this update was primarily backend. There were a few frontend changes, such as for optimization or just some cleanup/to-do items that were easy to address during a rewrite, but the intent was not to have some kinda drastic change in the user experience beyond faster load times.