Tell HN: Check out my app, RunMeLater, Web Service for CRON Jobs/Scheduled Tasks

I really like this idea! A few suggestions:

* Is there no way to delete a task? I wouldn't consider using it unless there were, because it would get awful annoying if I needed to change the schedule but was incapable of doing so.

* It would be nice if there were more options for the HTTP request (most especially sending POST parameters, but also stuff like HTTP auth would be useful)

* Perhaps nitpickey, but could you make the interface more RESTful? That is, POST to create an unnamed task, PUT to create/update a named task, DELETE to remove, GET to get information about a task.

If you're wanting to be more RESTful, you should be creating tasks with POST instead of GET, since you're creating new resources. How do you cancel tasks? Might I suggest a DELETE to a URL, specified in the return?

Thanks for the comments. I'll definitely improve on the documentation.

I decided to release it on HackerNews first because I know you guys would understand what I'm talking about.

I wanted a service like this when I was using a shared hosting account for my website. The host doesn't give you a lot of access to the servers and set up CRON jobs.

Let me design a better website for you. Not mindblowing, as I don't have much time, but better.

EDIT: For free, by the way.

You know what would be cool? If it could send me a summary of what my script told it. For example, I do batch sending of emails every 10 minutes. It should tell me at the end of the day: Sent 120 emails. 3 emails failed, they are max@max.con, max@max.ckom etc.

I still wouldn't pay for this - it's too easy for me to setup cron for me to pay you to do this for me. You'd have to offer me a lot more compelling reasons to pay you for this.

For ideas on how others have done this (including business model and pricing), check out http://www.webcron.org/.

Ah, this site takes me back to an online ballot box stuffing incident earlier in the year...

I've read a case study of someone with essentially this service trying to make money by offering a pro service. It's not going well.

So it might be hard to get enough people to pay for it to make it a sustainable business, but at the same time I would have a hard time recommending anyone do anything important through a free service where I couldn't see what the provider was getting out of it.

* Renders poorly on Firefox 2.something on Vista. Basically same issues as others have noted

* Use of the word "cron" threw me off. I was wondering how your Web service was going to interact with the cron service I have on my machines. But, of course, that's not what it does at all.

Suggestions: Fix the layout, and make it obvious what problem you are solving.

How do you stop people from using this to waste other peoples resources ?

I think with the market you would be aiming for you would want to make this as simple to use as possible. I'd ditch even really making much mention of CRON and make a solid and easy to use UI for setting up the tasks. Your really aiming for the crowd that uses basic shared hosting so the simpler the better with the web service taking a back seat.

As a business model possibly you could make viewing the results of jobs and stats visible on the website only for people that use the UI to create. People setting up through an API are less likely going to be able to be monetized anyway.

Really great idea. Starting with the API is smart. I hope you build it out and let me create an account and manage my tasks and check out logs, etc.

Paid Version: Launch Web requests based on phone calls. Use Twilio for the telephony and use your json results to make it more interactive.

I literally just implemented a slightly beefier version of this for a project I'm working on so obviously there's a need. There's a python implementation (dateutil) of the icalendar rrule format which handles nearly arbitrarily complicated repeating schemes with a decent API that you might want to look at.

Seems pretty cool actually. I'd break up the page and add some navigation. I'd recommend the jQuery UI Tabs plugin for this just because it's so fast. You could have a Home tab, FAQ tab, and Examples tab, perhaps?

Do you have a business model or is this just a resume bullet point?

I would prefer to have a panel, where I can access and manage my tasks. If you want to add the "delete" feature, you'll need accounts . Then a management panel will be very useful.

Minor note: cron is lowercase. If you're advertising to cron users, they might not trust something that misspells what it's supposed to replace.

a couple of thoughts regarding your app:

* unfortunately, the html does not render correctly in my chrome browser, e.g the first url on the left overlaps with the right column.

* depending on what you d like to do with it, you might want to consider to revamp a bit the UI?

* what's your intention with this service? Is it a side project or do you plan to commercialize it?

I like the idea of it, but you should really document this more. It's just all on one page now, and that makes it hard to read imo.

How does it work, and what is the point of such a webservice (your page gives no information concerning both) ?

hmm.. quite interesting. I wonder which scheduler did they use - cron based or roll-your-own using Erlang, Scala or something

Last year I was spending time searching for a CRON interface that could be subscribed to by iCal. I thought that it would be useful as a real-time correlation tool, allowing a a NOC to a calendar on a large screen and quickly gleam if a recent resource spike might be related to a cron entry (and which one). This might help with that.