And have a rake file in the www's repo to copy its content in assets.

So images for docs can be stored here.

This would make it easier for someone to read all documentation in a linear way or to download it for reading offline.

next and other commands need to have YAML flags to indicate they don't exist in other languages. They currently 404 because links in the language switcher are active.

rethinkdb.com/api used to have a lambda at the end of the command list to indicate the end of the API docs. This needs to be added back.

Add a recipe to the "Basic commands" section to get all documents for a table.

The node binary is named nodejs in Ubuntu.
Create a symbolic link works but @AtnNn mentionned something about nodejs-legacy which does that for you.

See rethinkdb/rethinkdb#1451 for the whole error.

There are a limited set of issues that can be presented in the UI, for example:

• Servers that should be removed from the cluster
• Metadata conflicts
• Redundancy issues
• Zombie servers

...and a few more. It would be useful to describe these different cases, and the consequences of resolving them (and would help simplify some error messages if we had a public resource for users to refer to).

Issues that are not documented:

• Dead machine
• Zombie machine
• Vector clock
• Unsatisfiable goals
• Log file not available
• Port conflict
• Name conflict

See rethinkdb/rethinkdb#933 for more details.

As requested by @jprichardson:

I wish there was comprehensive documentation on the RethinkDB configuration file. All I see is this: http://www.rethinkdb.com/docs/cluster-on-startup/ while, it's acceptable, it's not as good as it should be.

We should say that the primary key can be used as a secondary index too -- See rethinkdb/rethinkdb#1529 (comment)

We should also mention somewhere that we don't have an optimizer and that users should explicitly mention the index.

This was suggested by a user on HN a few releases back. Examples of limitations that would be useful to describe are:

• maximum document size
• maximum number of tables / databases
• maximum number of shards
• minimum disk space required

... to name a few.

Rewrite the generic build instructions to use git as a source instead of a tgz.

• API docs for eq_join, inner_join, and outer_join doesn't describe how to use zip to combine the output
• In the joins documentation, we should explain how to use pluck and map

The doc about r.literal is spread in filter and merge.

We should add a page about it (would it be only to help people google errors about r.literal).

http://www.rethinkdb.com/docs/cookbook/javascript/#filtering-based-on-multiple-fields

"You can also use the infix notation (passing all arguments to r.and), if that's what you prefer:"

If I'm not mistaken, the described notation is prefix notation not infix.

I'm not quite sure where this fits into your guys scheme but I think there should be a section in here somewhere to get driver developers started. One thing in particular that I think would be good to include is a rundown of optimizations we make in our drivers to get better performance such as converting to JSON and deserializing values lazily.

Previous issue rethinkdb/rethinkdb#952
Syntax rethinkdb/rethinkdb#884

We are missing the default argument in the syntax.

The docs about filter need more improvements, but let's keep this issue to just default (which is tricky enough).

We should write docs about these errors, and probably do the same for Python and Ruby too.

Each command should list its memory complexity.

Describing typical use cases for RethinkDB helps answer common questions like, "Is RethinkDB useful for tracking analytics?" or, "Can I use RethinkDB as an alternate data warehousing solution?"

Outside of dependencies, we should write unified build instructions that work across all platforms.

Assigning to @AtnNn -- we'll make this an include and simplify / abstract the install docs.

Does not require an explanation.

rethinkdb/rethinkdb#869 introduced these arguments to slice, between and delete_at. But in the API docs they are only documented for between.

See https://twitter.com/FaisalAbid/status/396713970767106048

We don't have any docs for rethinkdb proxy and it's actually a very useful feature that I've described to a number of people over IRC and had them be delighted. Let's add docs for it.

There are currently no docs on using the native Python date objects (see the tests for an example).

@kevincox said this in the original issue (rethinkdb/rethinkdb#620):

When selecting and designing with a database it is important to know the cost of
what queries in what situations. For example, it is often reasonable to use a
quadratic query, if you know the size of your dataset will be under a certain size.

It would be very useful to have the expected time complexities of the queries in the
documentation and the cases for which the complexity will be different. This is
something that Redis does really well for example.

This was met with complete agreement by everyone involved.

See rethinkdb/rethinkdb#1577

In JavaScript

r.connect(host, callback)

In Python and Ruby

r.connect(host)

host being a string.

Some changes have been made to not interrupt noreply writes
See rethinkdb/rethinkdb#1388

We should update the relevant API pages before releasing 1.11.0

There is currently no doc about r.any.

See rethinkdb/rethinkdb#1511 and rethinkdb/rethinkdb#1488 for more information.

This query doesn't behave as most of the people would expect

r.table("users").filter( lambda user:
    user.has_fields("age") | user["age"] == 0
).run(conn)

What they want is probably this

r.table("users").filter( lambda user:
    user.has_fields("age") | (user["age"] == 0)
).run(conn)

When we have an LTS release, we will need versioned API docs so that changes in semantics or additions / removals of commands don't confuse users.

Added with rethinkdb/rethinkdb#1329

The build instructions for each distribution need to be updated.

For ubuntu, that means adding libncurses5-dev and libtinfo-dev to the list of dependencies

As an example, brought up by @AtnNn in rethinkdb/rethinkdb#1061:

Saying that "with_fields is just has_fields followed by pluck" would be a lot more useful if has_fields and pluck were links.

Suggested by krz on IRC:

<krz> is there a getting started guide for rails applications?
<krz> you guys should really come up with a rails demo app using the ruby wrapper
<krz> just thought, if there was a guide up, just to get integrated with rails. that would be awesome

We have some links that point to the old API (like /api/#py:selecting_data-table) that we need to update.

Working on it now.

We should consider adding sample table data and the output of ReQL commands to every example.

This would likely require a widget with three tabs: example, sample data, and output. We should look at other complex APIs to find constructive examples.

We should add a page with all the types in RethinkDB, especially with stream/sequence/array/selection.

Original issue rethinkdb/rethinkdb#1337
Useful comments https://github.com/rethinkdb/rethinkdb-www/pull/85

See @jdoliner comment -- rethinkdb/rethinkdb#1529 (comment)

How to use secondary indexes for

• filtering
• ordering
• joins

Scrolling seems to cause artifacts.

Reported on IRC:

Kamilion: Hey guys -- Typo in the docsite. http://rethinkdb.com/docs/secondary-indexes/ <--- bottom of page, IndexdDrop and IndexdList... somehow an extra d slipped into both of those...

https://github.com/dancannon/gorethink

Being able to filter ReQL commands by the type they operate on would be useful.

@jdoliner suggests that we add examples that use functions to our secondary index docs.

http://rethinkdb.com/docs/sql-to-reql/

It's good, but could be made even better. Assigning to self.

At the bottom of http://rethinkdb.com/api/javascript/

Help us improve the docs by editing this document