I haven't tried any possible solutions. Building an inhouse monitor is a possibility, but I would pay for a service if it was only $25 pm. Particularly if it included the option of buying Sys Admin/DBA support, charged hourly as required, with the aim being the effective outsource of day to day monitoring and tech support of a site.
I really like your idea to buy sys admin/dba support as a topup when needed. I would love to hear what you think of New Relic. They do have a $24 a month option if you commit to a year. Also what are your thoughts on free solutions like monit or Nagios.
I recently came across http://amon.cx/ which does a few of the things you mentioned, and it's a one time purchase (with a free basic version) rather than a SaaS. Might be worth taking a look at.
I'm working on a simple version of this which I plan to open-source soon. You basically cron commands like df and uptime, pipe them to mail and send to app. App uses MailGun to read mail and then create reports.
I asked this question wanting to know what day to day issues people struggle with, what tasks they repeat a dozen times a day, what 20 process step can be reduced to 3 or even better completely automated.
From validation. It's a pain to write validation code and to do it cleanly, give me an ajax service that I can hand the validation off to. I just want to be able to say field x is a number, field y is an address, etc.
Hmm I think trying to convert such a task into a service will always introduce some language to describe your field and it's constraints and then you might as well write it yourself.
Also with such a critical part of your system, getting information from your user, if the service is slow or down you will run into troubles.
What specific programming languages and/or frameworks do you require a solution for? Please let me know if I missed the point completely. Thanks for your answer!
Well I was imagining some kind of ajax thing that validated fields on the fly as the user filled them in.
Validation is non-trivial for many fields. For example almost no-one implements email validation correctly as per the RFC because it's hard to get right, and most people just use a hacky regex that ends up accepting invalid emails and rejecting valid ones.
Or with credit card validation, many places don't validate credit card numbers (i.e doing the validation checksum, checking if the card is a mastercard that it's got the mastercard prefix) before sending them to the card processor site. If the card processing services rejects the card and the user has to re-enter the card details, chance are they won't bother and you've lost yourself a sale. If you validated for obvious mistakes on-the-fly (i.e user sees error immediately) you wouldn't lose as many of those customers.
You could also do things like supplementing the user supplied data, say the user supplies an address you could automatically geocode it and put it in a hidden field for the website.
So you mean client side validation correct? I mean to validate form submission that needs server side information, you would still need to build something on your own.
Yes. Normally a user filling in a form is the last step before conversion (signup, purchase, etc) so any users you lose at that step are horrendously expensive, good client side validation can make a significant difference.
Validation for the purpose of having clean data in your database is obviously important too, but less so. But obviously if you had a service for providing the first it wouldn't be hard to offer it as a server side services as well.
$25/month is kind of an ambitious price point for small teams or individuals. That's like implying that your service would be more valuable than Basecamp or Github.
Surely it depends on the problem being solved though? If I was making a new project management or source code hosting service I would have to think twice about charging the same or more than my competitors alternatively I need to compete on something other than price. Also if a you need a solution to problem X how can the price point be compared to an established company solving problem Y?
>if a you need a solution to problem X how can the price point be compared to an established company solving problem Y?
If problem X and problem Y both cost $25/month to solve, they should be roughly equal pain-wise. By starting at price point of $25/month, and working your way backwards to find problems, you should expect to be tackling Github- or Basecamp-sized problems.
I think that's non-sense. Just because X and Y cost the same to solve doesn't mean they are equal pain-wise. It doesn't take into account any of relevant factors like market size. I've seen a lot of services that are trivial in comparison charge a lot more because it's a much smaller market and/or the value of a very specific tool is worth far more for people in that niche.
X and Y can cost the same amount and solve different amounts of pain and both be priced accurately.
How?
Amount of Pain X Solves < Amount of Pain Y Solves
X solves a problem for a small market where the consumers are still find it worth $25 to solve the problem and have less options.
Y solves a massive problem that a lot of people have (big market). Prices have been driven to $25 by many competitors.
Different amounts of pain, different amounts of complexity, same price. Y's price has been driven down to 25 by competitive factors. X's price is higher because a lack of competitors. All this requires is that one is under-priced relative to value delivered but not to the market's competitive forces. Therefore neither is under priced given the reality of both sides of the market.
social network api where user can login via different networks and import contacts/friends/following. Must be able to have different types of users. and add extra fields like hobbies etc. on sign up form. Must have a javascript widgets I can drop on a screen and cusomtze look and feel or api I can call with backend. Tracking of user. User can share/post message to friends followers etc. mobile sdk also.
By automatic do you mean it should hook into your accounts and automatically categorize various items as expenses and provide an income statement and balance sheet every month?
Yep, as automatic as possible. It could try and auto categorize items and allow the user to fine tune the categorization. But basically doing what an accountant normally does automatically.
I'd want a SaaS that when given a URL of an article or piece of content I just wrote... gives me a bunch of related Twitter followers, potential LinkedIn Groups, bloggers, etc where I can pitch my content to. Content marketing tool, so to speak. Hard problem, but hey, that's why I'd pay a lot for it.
33 comments
[ 3.1 ms ] story [ 86.9 ms ] thread- Front end web site availability
- APIs available
- Back end processes are running
- Database up
- Disk space available
- Error logs for important issues
The database checks would include referencal integrity and sanity checks on volumes to warn of potential issues eg a failed module or abuse by a user.
Daily and weekly reports by Email, RSS, Twitter, web tool. SMS for emergencies requiring immediate action.
Option to have support from a Sys Admin/DBA automatically trigger to fix minor issues.
Looks like this: http://cl.ly/image/0Z15301Q3L1u
Currently frustrated that highrise does not timestamp tags, or allow tags to be sorted by a timestamp.
If you could add that feature as a plugin or via their api I would gladly pay $25 per month.
Haven't looked extensively for a solution, but it's certainly a pain point at the moment.
Thank you for the answers so far!
Also with such a critical part of your system, getting information from your user, if the service is slow or down you will run into troubles.
What specific programming languages and/or frameworks do you require a solution for? Please let me know if I missed the point completely. Thanks for your answer!
Validation is non-trivial for many fields. For example almost no-one implements email validation correctly as per the RFC because it's hard to get right, and most people just use a hacky regex that ends up accepting invalid emails and rejecting valid ones.
Or with credit card validation, many places don't validate credit card numbers (i.e doing the validation checksum, checking if the card is a mastercard that it's got the mastercard prefix) before sending them to the card processor site. If the card processing services rejects the card and the user has to re-enter the card details, chance are they won't bother and you've lost yourself a sale. If you validated for obvious mistakes on-the-fly (i.e user sees error immediately) you wouldn't lose as many of those customers.
You could also do things like supplementing the user supplied data, say the user supplies an address you could automatically geocode it and put it in a hidden field for the website.
Validation for the purpose of having clean data in your database is obviously important too, but less so. But obviously if you had a service for providing the first it wouldn't be hard to offer it as a server side services as well.
I'm curious, are you a developer?
If problem X and problem Y both cost $25/month to solve, they should be roughly equal pain-wise. By starting at price point of $25/month, and working your way backwards to find problems, you should expect to be tackling Github- or Basecamp-sized problems.
>I'm curious, are you a developer?
Yes, I'm a developer/consultant.
How?
Amount of Pain X Solves < Amount of Pain Y Solves
X solves a problem for a small market where the consumers are still find it worth $25 to solve the problem and have less options.
Y solves a massive problem that a lot of people have (big market). Prices have been driven to $25 by many competitors.
Different amounts of pain, different amounts of complexity, same price. Y's price has been driven down to 25 by competitive factors. X's price is higher because a lack of competitors. All this requires is that one is under-priced relative to value delivered but not to the market's competitive forces. Therefore neither is under priced given the reality of both sides of the market.
should be secure, auto detect formats, etc.