Hacker News Reader

Comment by rb2k_ | Hacker News Reader

It seems like a really nice CMS. I love aloha as an editor and the native multi site capabilities.

But sadly there is no demo link available (crossed out) and the installation is quite a bit more complex than for the usual Rails CMS :(

It's not too bad, but usually you just do a gem install and then there's a new helper method allowing you to setup a matching project.

For some very strange reason, I get a segfault when trying to do a "bundle install" in the process described in the documentation.

View on HN

Hi @rb2k_, thanks for the feedback. About the demo link, since we do not have a stable version of the engine yet, we chose to disable our demo platform. But it'll be coming soon !

You're totally right about the installation procedure. Like the other open source projects such as Spree for instance, we miss an executable which will help users to start a new engine with ease. That could be perhaps your first contribution to the project :-)

Finally, concerning your segfault error, if you have time, please open an issue in github (https://github.com/locomotivecms/engine/issues) or on our mailing-list (https://groups.google.com/forum/?fromgroups#!forum/locomotiv...)

thanks again !

I also almost left without looking closely because of the disabled demo link.

I look forward to the demo, because from my subsequent cursory glance, this may be exactly what I need for a non-profit project I'm working on.

https://github.com/locomotivecms/engine/issues/369

thanks @rb2k_ !

I would be using it if it didn't insist on MongoDB.

I just dont trust data I care about to it

with a CMS you have barely any writes. Not sure your fear is well grounded in reality :)

I actually agree with the OP in general. I have a mistrust of MongoDB overall, and avoid using it in big, data-intensive situations...

BUT, in LocomotiveCMS and a few others, it make brilliant sense. Like you mentioned, it's very lightweight in terms of DB-writes, and allows for a flexibility that when put together the way the the Locomotive team has done makes a CMS a wonderful thing for end users.

So I think you guys are both right in the this case. Choose the right tool for the job and all that... and the Locomotive team definitely chose the right tool.

Doesn't everytime you update anything with the cms require a write? (asking purely out of ignorance to learn not to be snarky)

Not to mention everytime a user of the site updates content

yeah, but that's about write every minute or so, even if you have 100 monkeys on 100 typewriters

Just an opinion - none of us will be able to push Locomotive in an enterprise setting (which is precisely where CMS makes a lot of sense) without it supporting a relational DB.

Dont get me wrong - I get the value proposition of mongo - but in this case, it is taking away a lot and not giving me anything in return.

We chose the tools that worked best for us and for our clients. LocomotiveCMS is already used by startups, web agencies, and big companies to design, deploy, and update sites every day. If Mongodb becomes a limiting factor, or if someone provides an equally elegant implementation on top of a RDMS, we'll happily switch. Until then, we'll focus on building our vision of the most elegant and friendly CMS possible.

could you elaborate on "equally elegant implementation on top of a RDMS" ? Genuinely want to know what advantage did mongo give you in a CMS setting.

Were you able to design a better architecture because of Mongo/nosql or was it developer productivity ?

I'm not well-versed in Mongo, but it seems like the comment was saying that he didn't trust Mongo to handle any writes. Your response seems to say that since there aren't many writes to lose, it's okay. Does that concede that Mongo will lose data? (I didn't know that about Mongo.)

To explain the parent comment, in a write heavy application, you would probably want to turn off transactions in Mongo in return for the performance gain.

What the comment you're responding to is saying is that when you aren't making many writes, you can afford to switch transactions on and have Mongo lock the whole database because the odds of having a read come in at the same moment are much lower than if you were making a write every second.

Ahh, thanks for the explanation. This clears up something I didn't know about Mongo (next to nothing other than that the other commenters all seemed to indicate that it should not be used for writes).

Much more informative than a downvote.

Reply on news.ycombinator.com