- cross-posted to:
- hackernews
- [email protected]
- cross-posted to:
- hackernews
- [email protected]
You must log in or register to comment.
Not as drastic as the headline makes it out to be, or at least so they claim.
“We acquired Tumblr to benefit from its differences and strengths, not to water it down. We love Tumblr’s streamlined posting experience and its current product direction,” the post explained. “We’re not changing that. We’re talking about running Tumblr’s backend on WordPress. You won’t even notice a difference from the outside,” it noted.
We’ll see how that actually works out. Tumblr’s backend has always seemed rather… makeshift, so I’m curious to see how they manage to do that. Given Tumblr’s technical eccentricities, a backend migration could probably do a lot of good for the functionality of the site, if done properly. I have my doubts that WordPress’ engineers will be given the time and resources to do a full overhaul/refactor though, so I’m fully expecting even more janky, barely functional code stapling the two systems together.
WordPress is built on decades of hacky code, probably more so than Tumblr. I would be shocked if this is an improvement.
is it decades of hacky code, or decades of battle tested code?
I haven’t touched wordpress in… many years, but I’ve seen far too many developers look at old code and call it junk… only to break things horrifically when they attempt a rewrite.
Hacky.
Wordpress has a reputation for the most moronic security issues. Especially when it’s built on PHP, which has its own reputation for moronic security issues. And that’s saying nothing about the quality of plugin developers or plugin code.
I’ve worked on Wordpress sites, plugins, and themes. That was many years ago now, but I doubt it’s changed that much. If anything, it’s mostly benefited from improvements to PHP.
Has to rank as one of the most exploited pieces of software ever.
Definitely be not aided by the fact it’s targeting an audience without the skills or knowledge to adequately configure, maintain and monitor it. And the plugin community only makes the vulnerability exposure worse.
Yup. I imagine a lot of users install a lot of plugins they don’t actually need, which just expands the attack surface.
Kind of the old Windows vs Mac problem though. It gets so many exploits because it is so ridiculously popular. No one is going to bother looking for exploits in shit that no one uses right? I’m sure they’ve got problems like any project but I’m not convinced they’re THAT bad. Not to mention a lot of exploits you see are plugins doing dumb shit, not WP itself.
Both honestly. Very spaghetti, but noone can deny that it just works from a user perspective. Would I want to maintain the code? Hell no! Do use it as an end user? Hell yeah!
Nah, not touching that with a 10’ pole. There have been far too many exploits for me to feel comfortable putting any of my important data on it. And it’s not just that it’s popular, the level of sophistication for these attacks are… alarmingly low.
It’s a public site that’ll be backed up regularly, what kind of important data would you be putting out publicly?
If it’s an e-commerce site, than people’s payment info, name, and address. If it has a login, then their login information (which they’re most likely reusing elsewhere). Even if it’s just a static site, than any data that might be hosted on the same server.
my thoughts exactly. Who in their sane mind sees WordPress as a solid foundation for anything?
you must be truly desperate to come to me for help.
LokiWPMost large publishing companies, the white house and various government departments all use WordPress for their main sites. Its the third party integrations that cause security issues, not the core code.
Yet the third party integrations are pretty much the whole point of WordPress.
Indeed, but using poor ones or not keeping them updated is what causes the wrong opinion that WordPress isn’t solid.
30% of the most popular 1000 websites are built on WordPress supposedly.
Sure, and who is vetting the plugins? How often are unmaintained plugins replaced in those popular websites? How quickly are vulnerabilities patched and applied?
The whole thing is easy to set up, but unlikely to be properly maintained.
Not as familiar with WordPress, but if that’s the case, yeah, I don’t have high hopes for this going well…
Every comment in this thread might as well be hearsay. I wouldn’t take it too seriously. I think I’ll trust the corporation that runs wordpress.com and maintains the open source WordPress project instead to know what they’re doing with WordPress.
Wordpress supports activitypub tho, so that could be cool if they want it to be.
I read a couple of Tumblr blogs. If I could follow them from Mastodon instead I could delete that app entirely.
Use rss feeds maybe? Adding
/rssto a tumblr blog’s url (in the x.tumblr.com format) shows an rss feedI’ll believe it when I see it.
They summarised the history of Tumblr, but failed to mention how they lost 3 quarters of their users by banning porn?
Like, two owners ago. Wordpress took Tumblr off Verizon’s hands for $3 million USD, ~six years after Yahoo! bought it for $1.1 billion.
Where the fuck does yahoo even get money from to do this kind of shit at this point?
Rich people can always get money
The people can, but companies still need some kind of income to exist. The owners/ceos will just golden parachute away from the corpse
In order to tangibly pay employees/rent/servers a company needs either profits, subsidies, or
a ponzi schemeinflated stocks.
yahoo
Nowadays, I don’t know that they could, but more than a decade ago they still had enough mail and search users to be somewhat relevant, and Marissa Meyer had just taken over after she left Google. There was a real thought that Yahoo! could so something new. It obviously didn’t pan out, but for a hot minute, people really talked about Yahoo!
Even after Automattic acquired it, the site continued to lose money at a rate of $30 million each year, the company’s CEO Matt Mullenweg had said.
I still wanna know what they’re spending all that money on, because I’m sure it’s not developers or even servers. The idea that they can only be profitable if they’re constantly growing their user numbers is an investor idea that’s doomed to fail eventually and why so many social media sites are crashing right now
if they’re constantly growing their user numbers
All social media needs to constantly grow because attrition. Social media requires basic levels of user ship to be functional, even lemmy. Its a network effect where you need to have certain levels of users for some emergent properties to exist. For example, I speculate that defederation early between .ml and .world was the trigger that will eventually kill lemmy, principally because this results in fragmentation and a reduction in the properties we would get from “more users”. Having more users begets more users, more content, more memes, etc. And I don’t necesssarily see the defederation as something unneccessary, but what I’m describing is an inherent property of networks. Its not something that can really be argued with because this behavior is consistent across physical, biological, social networks. It just “is” as a property.
So foundationally, you can’t sit still on a train moving backwords (which it always is). An organism needs to be constantly recruiting and growing new cells into its network because its also always dying. Growth is “holding still” for any networked system.
Lemmy can work without user accounts made directly on the server. I’m posting from a completely different instance using completely different software (mbin), and I can see both .ml and .world and interact with them both just fine despite their defederation from each other.
It’s kind of more like a random street gathering instead of an exclusive club.
Or like the golden age of instant messengers, where you had multiple choices of multi-client apps like Trillian.
You still had individual accounts for each IM platform, but a single app to chat on any platform.
Related: I’m a big fan of Beeper, and they were recently acquired by Wordpress too.
I see no mention of ActivityPub in the article, but I’m wondering if this is part of their plan to eventually integrate Tumblr into the fediverse as well.
However I agree with others that this will likely result in hella janky hackable websites first…hopefully it smoothes out.
yeah i remember when that article came out in 2023
Funny how all these social media corporations were run as ponzi schemes…
deleted by creator
Ah yes, WordPress, renowned for it’s robust social engagement tools.
It’s the kind of decision you announce over Zoom so people don’t riot
https://wordpress.org/plugins/activitypub/
Hopefully they’re planning on using that to integrate with the Fediverse.
Neocities already has its own internal engagement stuff (with people following each other, commenting on updates etc), it probably wouldn’t be too hard to throw that into ActivityPub.
Just like wordpress though neoticies is much more “here is my stuff, browse through it” oriented than tumblr, which is at least 70% towards twitteresque “here’s a firehose of different stuff of different people please make comments and retweet every image”.
Don’t envy anyone involved in this… it’s a lot of work
…and the WordPress codebase is utterly horrible. I don’t envy them at all.
I don’t know what you mean? How could a codebase so often patched for exploits and problems be so horrible, surely it’s a shining example of all that is good with the internet? ;-)
I mean it’s probably the codebase most targeted, so it makes sense
I was supporting a WordPress site and we’ve had issues getting blacklisted by Internet providers because some WP scripts were in the malware database (local file matching WP github exactly). What a nightmare.
Seeing how much CPU and memory is using a single WordPress blog, i wonder how much it will cost to host half a billion WordPress blogs
My guess is it scales a lot more efficiently the more you add. Still probably costs a lot though.
Most of them probably get 0 visits or close to it.
I mean when I made one such blog, the default install did not have a single index in the database? Like even a small dataset took seconds to process.
People love to shit on PHP, and I also love to shit on WordPress, but performance was definitely possible to easily improve.
wtf i don’t want this i want the blog ive had for a decade on tumblr
They’re not actually going to change anything you can notice. I guess they’re just changing the back end for… Reasons
Sounds stupid. I wonder if this makes it easier to sell the content to AI scrapers?
It makes sense that they’d want to move it to a single codebase rather than have both Wordpress code and Tumblr code in the same organization.
Anyone else old enough to remember when Wordpress was called b2? Good times.
Tumblr’s codebase is also both quite old and infamously terrible, even if it’s from being shuffled around companies a bunch.
Centralising its backend into one platform doesn’t like too bad of an idea.
So much WordPress hate but I survived off making WordPress sites out of college. The new built in site theme editor is great. No more hacky plugins.
Yeah it was a nice tool for making sites fast. I loved using it but haven’t in a very long time. I hope it’s still nice to work with like it was before. Even with PHP it wasn’t that bad of an experience lol.
The overuse of plugins is what made it a mess. But you need less and less nowadays!
Automattic says the move to WordPress will have its advantages, as it will make it easier to share the company’s work across the two platforms.
I foresee no issues
Wordpress is just the worst
The open-source software from WordPress.org is great for blogs. It becomes the worst when you try it make it do more than that. Even worse is WordPress.com which is very different and uses a very locked-down and restricted proprietary version of the WordPress software. They charge $25/mo for the tier that lets you add custom CSS.
Additionally, Automattic gets a free pass of violating the WordPress terms of use for the WordPress name and logo to intentionally trick people into thinking the paid platform at WordPress.com is the same as the free and open-source software from WordPress.org. They get to leverage the non-profit’s name and likeness and gets preferential treatment to funnel business to their for-profit company.
Wow. For real, I always just assumed that .com was the commercial arm of .org. Holy shit.
Edit: So, for anyone curious, .com is owned by Automattic, who also own Tumblr, Beeper, PocketCasts and Buddy Press. The WordPress project and .org are owned by the WordPress Foundation. Automattic makes some contributions to the WordPress project but they and the WP Foundation are seperate.
Completely separate entities besides the owner being the same.
I think you misunderstood the comment you are replying to.
The WordPress Foundation does not have the same owner as WordPress.com.
Matt Mullenweg effectively is the ‘owner’ of both, yes. Though the term ‘owner’ isn’t necessarily precise, it effectively conveys the idea.
Ah sorry, it sounded like you were reacting to the comment you replied to, which it was more like you were adding information.
The face I made when I read that title and it was not good.
Ewww
~~Wait wtf. Automattic was sold in a fire sale?
When? Why? How?They were not making money? I thought they had WordPress.com - that should be making pretty good right? Or paid plugins like jetpack - those gotta be making money.
How did this happen that they got sold in fire sale?~~Edit oh. Ive read it wrong. Tumblr was sold in a fire sale, okok. Yeah I could see that.
Hopefully they’ll at least use Bedrock and not default Wordpress.
Or modify it themselves.
Or use a shitload of insecure plugins for wp to do what it wasn’t made for.
Who knows what they use under the hood but WordPress architecture is horrible
