5 ms·

My sister in law uses word press for her small online art store. It's... painful. I've had to help her fix a completely broken site at least a couple of times.

by kayson 3y ago

My sister in law uses word press for her small online art store. It's... painful. I've had to help her fix a completely broken site at least a couple of times. It mostly seems to come down to the fact that if you want anything beyond a simple blog, you have to have plugins. Tons and tons of plugins. Once you go down that road it's a huge mess of compatibility issues and poor to nonexistent support. It seems like you can get better help when you pay for things, but even then, in one case the devs were completely silent. I couldn't find any alternative store fronts that weren't way beyond her near-zero budget but if anyone has recommendations please let me know.

paulryanrogers 3y ago
IME it helps to keep the number of plugins limited and judicious. Stick to very popular plugins or those with a sustainable business. Basically big enough it's certain to be forked if the original authors stop supporting it.
- AussieWog93 3y ago
  Exactly this. My website was hacked a while back, when I was installing random disreputable plugins. Remade it but thought before I installed and suddenly the site is performant and secure.
mikepurvis 3y ago
Wouldn't Shopify be a pretty strong candidate for that kind of thing these days?
- kayson 3y ago
  Yes but it's just way too expensive for her sales volume.
- sumoboy 3y ago
  People just default to Shopify like it's some sort of miracle, everything about it beyond out of the box is PIA. Tons of add-on's (aka. plugins) that all cost to extend the basics like forms, seo, and fulfillment/shipping among the typical. The costs add up even for a basic store and theme dev requires some knowledge. Wordpress gutenberg is like a bad experiment that has no end game. They have spent years developing something is incrementally improves at snails pace. Wordpress is just plain and simple super accessible from a strong ecosystem of plugins, php is simple and widely known, it's an easy platform to host. The problem is all the alternatives like jamstack generators, headless cms, and even better CMS tools like craft just don't have enough popularity to leverage away from WP. I could easily spin up a woocommerce site just as fast as Shopify, but in reality not a fan of either. From an agency perspective, most clients could give a rats ass about things technically, they just want something working.
RadixDLT 3y ago
try drupal
- loloquwowndueo 3y ago
  Well if the issue they have with Postgres is complexity and dealing with plugins, Drupal is not going to help. It’s way more complicated and IIRC it’s plug-ins all the way down.
- ceejayoz 3y ago
  Drupal has the exact same problem.
- codecutter 3y ago
  Drupal has a steep learning curve. My personal website is on Drupal and now it is a nightmare even to upgrade it. I wouldn't recommend Drupal unless you have technical skills. It is not plug-and-play.
- karlshea 3y ago
  I've been making Drupal websites for 15 years, and especially lately now that its code is modernized wouldn't touch WordPress for any reason... but simple Drupal is not. It can be very simple to administer if the site is built right, but someone has to get it there and that takes some budget. Anyone in OP's situation I push towards Shopify/Squarespace.
- nevster 3y ago
  I once ran a Drupal site. Soooooo much more complicated than Wordpress. Definitely not recommended in this case.
joshspankit 3y ago
WP store front plugins are mature at this point and really focused on people already netting at least a few thousand dollars a month. Most of the “just starting out” crowd (and a decent chunk of everyone above that) went with platforms specifically focused on selling (Shopify, Squarespace, Square, etc)
waboremo 3y ago
>near-zero You aren't going to get better than Wordpress for that price. Closest options are Shopify and Squarespace. Both of them definitely not 0, with Shopify's starter coming in at around ~5 with the next jump up being 50/m, and Squarespace commerce at around ~20. Worth the price of admission for how much easier they are to use. The plus to both of these are that it's much easier to start adding different features to incentivize purchases and return customers. If the price is seriously a concern, it's advisable instead to help your sister in law move away from a dedicated platform and instead utilize Etsy, Instagram (Shopping/Shops), and Gumroad to their advantages, and the cost there is 0 (until you sell things then they take a slice of the sale). Several reasons for this beyond cost, your sister in law is likely already using platforms like Instagram for social reach, and most people are going to trust platforms like Etsy over individual sites if they aren't built well. So more easy sales. Once she's grown past a certain stage she can then fully hire someone to manage a custom solution, but until then do not let her underestimate how great Etsy, Gumroad, etc are!
- johndhi 3y ago
  I don't hate google sites but haven't done a store with it.
- ehnto 3y ago
  Worth noting, Shopify also takes a cut of sales, plus payment provider cut, and tight margin businesses might struggle on Shopify. I highly recommend it as a platform though, just make sure the business maths works.
rectang 3y ago
> you have to have plugins. Tons and tons of plugins. Once you go down that road it's a huge mess of compatibility issues and poor to nonexistent support. And security vulnerabilities. Core WordPress is fairly well hardened because it gets a ton of attention. Third-party plug-ins, not so much.
- chrismorgan 3y ago
  Actually, I blame WordPress even for most plugins’ security vulnerabilities, because its chosen architecture makes them much more exploitable than they should generally be. There’s also the problem that WordPress builds upon and relies heavily on being able to modify its own code. Although this does allow you to reliably do things like unattended automatic updates, which would probably be net good (normally improving things, though occasionally harming), it also means that even the slightest security vulnerability tends to become immediate total server takeover. This is why I blame WordPress core architecture more than individual plugins: in most ecosystems, most security bugs apparent in plugins wouldn’t be so exploitable. This design decision is probably responsible for most WordPress site hijackings (… which are very common). Taking all things into consideration, I suspect that going read-only file system and not supporting installing/updating/removing any code via its web interface would probably be good for WordPress’s security, despite it meaning most sites would never be updated—but it would certainly harm its ease of use, and I can easily see why it’s basically a non-starter. By contrast, choosing another PHP system as an example (since it’s one of the few popular languages that particularly supports this kind of self-modification), last time I dealt with Drupal (back in 5 and 6 days), it would actively complain if it could write to its own directory, urging you to make all but the directory it uploaded user files to read-only to the web server user account. Updates were then done out-of-band, using your user account on the machine which had write access to replace the files, rather than the web server’s account. (There were manual steps involving the web interface and local file operations, or tools like Drush to make it a single command.) Bad for ease-of-use, excellent for security. I haven’t had much to do with WordPress, but I gather storing executable PHP code in the database is also fairly common, which is almost worse than having a writeable file system. My vague recollection of Drupal is that although it supported “PHP code” as a content type, you had to turn it on in the config file, where it warned you of the vulnerabilities this opened. (My qualifications in this comment: I’ve helped recover several hijacked WordPress instances, mostly around 2015 and 2022; migrated one business off WordPress in 2022; done almost nothing else ever with WordPress; worked fairly extensively with Drupal 5 and 6 on a few sites quite a few years ago; worked a fair bit with Django over the years, and some in other web frameworky things in Node.js and Rust; haven’t done anything serious with PHP for years now, preferring Python, JavaScript and Rust. Certainly don’t trust me for any insights into WordPress. I could easily have made errors in this comment.)
tambourine_man 3y ago
That’s the problem people have with WordPress, the tons of plugins they install. Not much WordPress’s fault, aside from making writing and installing plugins easy and maybe not including a bazillion features few will use out of the box. I think Gutenberg is still not ready, but when it matures, it may be a better alternative to the page builder plugins like Elementor, etc. Comparing a hosted open source solution to Squarespace and the likes is non sequitur, IMO
- CSSer 3y ago
  I find it pretty funny that Gutenberg is now the fusion of the Wordpress ecosystem.
  - tambourine_man 3y ago
    If only there was as much progress in latter in so little time.
- mercer 3y ago
  > Comparing a hosted open source solution to Squarespace and the likes is non sequitur, IMO how so?
  - tambourine_man 3y ago
    You can change it however you want and be sure it will be running in decades to come. They’re not in the same category, to me.
- acomjean 3y ago
  I wonder how is Gutenberg not ready? We’ve been using it for 3+ years. It can be a little clunky but graphic designers and content creators like it for simple page/post layouts. It’s not always easy but it always seems logical.
paulpauper 3y ago
Wordpress is a full-featured CMS. Most people would be better off with something like a static site generator
- AnonC 3y ago
  I disagree. Most people want a GUI or browser based tool that works as a CMS (even if a limited one). They just wouldn’t be able to handle the 99.9% of static site generators that work on the command line and need the user to learn specific templating languages, image linking, tags, etc. People who drift to WordPress also want the cost to be lower and to be able to host almost anywhere (meaning, almost any web host should support it). The cost and usability combination hasn’t been solved well by tools other than WordPress. If there are any, I’d like to know or re-examine them.
  - marc_io 3y ago
    Have you seen Publii? It seems to be a solution somewhat similar to the one you described. It's quite limited compared to WordPress, which can be a good thing.
testbjjl 3y ago
> It mostly seems to come down to the fact that if you want anything beyond a simple blog, you have to have plugins. Tons and tons of plugins. Once you go down that road it's a huge mess of compatibility issues and poor to nonexistent support. Sounds like the libraries and open source software I work with everyday, to me.
arp242 3y ago
I'd probably just write it myself; unless there are very peculiar requirements a "small online art store" is not that hard or time-consuming to implement, and seems less effort than dealing with off-the-shelf software.

by kayson 3y ago

paulryanrogers 3y ago

IME it helps to keep the number of plugins limited and judicious. Stick to very popular plugins or those with a sustainable business. Basically big enough it's certain to be forked if the original authors stop supporting it.

AussieWog93 3y ago
Exactly this. My website was hacked a while back, when I was installing random disreputable plugins. Remade it but thought before I installed and suddenly the site is performant and secure.

mikepurvis 3y ago

Wouldn't Shopify be a pretty strong candidate for that kind of thing these days?

kayson 3y ago
Yes but it's just way too expensive for her sales volume.
sumoboy 3y ago
People just default to Shopify like it's some sort of miracle, everything about it beyond out of the box is PIA. Tons of add-on's (aka. plugins) that all cost to extend the basics like forms, seo, and fulfillment/shipping among the typical. The costs add up even for a basic store and theme dev requires some knowledge. Wordpress gutenberg is like a bad experiment that has no end game. They have spent years developing something is incrementally improves at snails pace. Wordpress is just plain and simple super accessible from a strong ecosystem of plugins, php is simple and widely known, it's an easy platform to host. The problem is all the alternatives like jamstack generators, headless cms, and even better CMS tools like craft just don't have enough popularity to leverage away from WP. I could easily spin up a woocommerce site just as fast as Shopify, but in reality not a fan of either. From an agency perspective, most clients could give a rats ass about things technically, they just want something working.

RadixDLT 3y ago

try drupal

loloquwowndueo 3y ago
Well if the issue they have with Postgres is complexity and dealing with plugins, Drupal is not going to help. It’s way more complicated and IIRC it’s plug-ins all the way down.
ceejayoz 3y ago
Drupal has the exact same problem.
codecutter 3y ago
Drupal has a steep learning curve. My personal website is on Drupal and now it is a nightmare even to upgrade it. I wouldn't recommend Drupal unless you have technical skills. It is not plug-and-play.
karlshea 3y ago
I've been making Drupal websites for 15 years, and especially lately now that its code is modernized wouldn't touch WordPress for any reason... but simple Drupal is not. It can be very simple to administer if the site is built right, but someone has to get it there and that takes some budget. Anyone in OP's situation I push towards Shopify/Squarespace.
nevster 3y ago
I once ran a Drupal site. Soooooo much more complicated than Wordpress. Definitely not recommended in this case.

joshspankit 3y ago

WP store front plugins are mature at this point and really focused on people already netting at least a few thousand dollars a month. Most of the “just starting out” crowd (and a decent chunk of everyone above that) went with platforms specifically focused on selling (Shopify, Squarespace, Square, etc)

waboremo 3y ago

>near-zero You aren't going to get better than Wordpress for that price. Closest options are Shopify and Squarespace. Both of them definitely not 0, with Shopify's starter coming in at around ~5 with the next jump up being 50/m, and Squarespace commerce at around ~20. Worth the price of admission for how much easier they are to use. The plus to both of these are that it's much easier to start adding different features to incentivize purchases and return customers. If the price is seriously a concern, it's advisable instead to help your sister in law move away from a dedicated platform and instead utilize Etsy, Instagram (Shopping/Shops), and Gumroad to their advantages, and the cost there is 0 (until you sell things then they take a slice of the sale). Several reasons for this beyond cost, your sister in law is likely already using platforms like Instagram for social reach, and most people are going to trust platforms like Etsy over individual sites if they aren't built well. So more easy sales. Once she's grown past a certain stage she can then fully hire someone to manage a custom solution, but until then do not let her underestimate how great Etsy, Gumroad, etc are!

johndhi 3y ago
I don't hate google sites but haven't done a store with it.
ehnto 3y ago
Worth noting, Shopify also takes a cut of sales, plus payment provider cut, and tight margin businesses might struggle on Shopify. I highly recommend it as a platform though, just make sure the business maths works.

rectang 3y ago

> you have to have plugins. Tons and tons of plugins. Once you go down that road it's a huge mess of compatibility issues and poor to nonexistent support. And security vulnerabilities. Core WordPress is fairly well hardened because it gets a ton of attention. Third-party plug-ins, not so much.

chrismorgan 3y ago
Actually, I blame WordPress even for most plugins’ security vulnerabilities, because its chosen architecture makes them much more exploitable than they should generally be. There’s also the problem that WordPress builds upon and relies heavily on being able to modify its own code. Although this does allow you to reliably do things like unattended automatic updates, which would probably be net good (normally improving things, though occasionally harming), it also means that even the slightest security vulnerability tends to become immediate total server takeover. This is why I blame WordPress core architecture more than individual plugins: in most ecosystems, most security bugs apparent in plugins wouldn’t be so exploitable. This design decision is probably responsible for most WordPress site hijackings (… which are very common). Taking all things into consideration, I suspect that going read-only file system and not supporting installing/updating/removing any code via its web interface would probably be good for WordPress’s security, despite it meaning most sites would never be updated—but it would certainly harm its ease of use, and I can easily see why it’s basically a non-starter. By contrast, choosing another PHP system as an example (since it’s one of the few popular languages that particularly supports this kind of self-modification), last time I dealt with Drupal (back in 5 and 6 days), it would actively complain if it could write to its own directory, urging you to make all but the directory it uploaded user files to read-only to the web server user account. Updates were then done out-of-band, using your user account on the machine which had write access to replace the files, rather than the web server’s account. (There were manual steps involving the web interface and local file operations, or tools like Drush to make it a single command.) Bad for ease-of-use, excellent for security. I haven’t had much to do with WordPress, but I gather storing executable PHP code in the database is also fairly common, which is almost worse than having a writeable file system. My vague recollection of Drupal is that although it supported “PHP code” as a content type, you had to turn it on in the config file, where it warned you of the vulnerabilities this opened. (My qualifications in this comment: I’ve helped recover several hijacked WordPress instances, mostly around 2015 and 2022; migrated one business off WordPress in 2022; done almost nothing else ever with WordPress; worked fairly extensively with Drupal 5 and 6 on a few sites quite a few years ago; worked a fair bit with Django over the years, and some in other web frameworky things in Node.js and Rust; haven’t done anything serious with PHP for years now, preferring Python, JavaScript and Rust. Certainly don’t trust me for any insights into WordPress. I could easily have made errors in this comment.)

tambourine_man 3y ago

That’s the problem people have with WordPress, the tons of plugins they install. Not much WordPress’s fault, aside from making writing and installing plugins easy and maybe not including a bazillion features few will use out of the box. I think Gutenberg is still not ready, but when it matures, it may be a better alternative to the page builder plugins like Elementor, etc. Comparing a hosted open source solution to Squarespace and the likes is non sequitur, IMO

CSSer 3y ago
I find it pretty funny that Gutenberg is now the fusion of the Wordpress ecosystem.
- tambourine_man 3y ago
  If only there was as much progress in latter in so little time.
mercer 3y ago
> Comparing a hosted open source solution to Squarespace and the likes is non sequitur, IMO how so?
- tambourine_man 3y ago
  You can change it however you want and be sure it will be running in decades to come. They’re not in the same category, to me.
acomjean 3y ago
I wonder how is Gutenberg not ready? We’ve been using it for 3+ years. It can be a little clunky but graphic designers and content creators like it for simple page/post layouts. It’s not always easy but it always seems logical.

paulpauper 3y ago

Wordpress is a full-featured CMS. Most people would be better off with something like a static site generator

AnonC 3y ago
I disagree. Most people want a GUI or browser based tool that works as a CMS (even if a limited one). They just wouldn’t be able to handle the 99.9% of static site generators that work on the command line and need the user to learn specific templating languages, image linking, tags, etc. People who drift to WordPress also want the cost to be lower and to be able to host almost anywhere (meaning, almost any web host should support it). The cost and usability combination hasn’t been solved well by tools other than WordPress. If there are any, I’d like to know or re-examine them.
- marc_io 3y ago
  Have you seen Publii? It seems to be a solution somewhat similar to the one you described. It's quite limited compared to WordPress, which can be a good thing.

testbjjl 3y ago

> It mostly seems to come down to the fact that if you want anything beyond a simple blog, you have to have plugins. Tons and tons of plugins. Once you go down that road it's a huge mess of compatibility issues and poor to nonexistent support. Sounds like the libraries and open source software I work with everyday, to me.

arp242 3y ago

I'd probably just write it myself; unless there are very peculiar requirements a "small online art store" is not that hard or time-consuming to implement, and seems less effort than dealing with off-the-shelf software.