4 ms·
Ask HN: What are you willing to cut for an MVP/first version running on the web?
Testing? CUT!
Security? CUT!
Anything but the barest bones features? CUT!
Backups? CUT!
Monitoring? CUT!
An attractive user interface? CUT!
Paid offerings? CUT!
User documentation? CUT!
What are you willing to cut to get the first version of your software up and running on the Internet?
- aaronbrethorst 11y ago> Paid offerings? CUT! Whoa whoa whoa whoa. No. My MVPs are essentially a Bootstrap UI wrapped around user accounts and the Stripe SDK, plus as little extra code as I can write.
- hoodoof 11y agoHave you launched many times on that formula?
- viraptor 11y agoI'm not sure how much you want to cut from each, but here are my reactions to some points: Security: What do you mean? Full audit: not useful. Getting basic TLS cert, copying default-secure nginx config from previous service, not intentionally breaking things your framework provides with an on/off switch. Usually that takes less time than setting up your own solutions. Backups: If you have any user-created data, "crontab; tar -zcf ...; s3cmd sync ..." - this is literally 5 minutes of work to set up primitive backup that will save you when stuff is on firew. Monitoring: Pingdom is free for a month - enough before you need actual monitoring. Those 3 points can be done to an MVP level in ~20 minutes. Why cut them?
- jeffmould 11y agoI take it these are things you choose to elect and not things you are saying someone should cut. What can and cannot be cut from an MVP/first version is going to depend greatly on the type of product you are launching in general. Using your list here are my opinions on each item: 1. Testing - in depth testing can be reduced greatly, but at least some minimal testing should be conducted. Especially if you have users signing up, taking user input, or processing payments. 2. Security - again, this one is going to vary greatly on what the product is exactly. 3. All but barest bone features - goes for any MVP 4. Backups - backups are cheap and easy to setup. No reason you can't do at least a minimal back of your data. 5. Monitoring - simple and basic setup, even using something like PagerDuty or Pingdom. 6. An attractive user interface - agree with you here. 7. Paid offerings - I used to be a strong believer in launch free plans to start and add paid offerings later. But if your eventual business model is to offer paid plans then you should start with something from the get go. Setting up Stripe is simple and you can be up and running within a day. 8. User documentation - cut away and add later. A FAQ section is not frequently asked questions if you don't have any users frequently asking questions :) So to answer your question, what I am willing to cut, or how much of something I willing to cut, all depends on the specific project.