Onto

checkout

Here's what 20 seconds of googling turned up: University of Rochester Grammar Style Guide oh hey look, a stackoverflow thread The truth about grammar: bailout versus bail out and there are so many more...

I think the conjugation is particularly helpful to see why it should be two separate words: "log in" -> "logging in" -> "logged in"

I have to mark up the pages

mark up

source | edit | rollback | link

set up

The verb set up, on the other hand, is usually found as an open compound (two words, no hyphen) in both American and British English.

It’s a “bug” and you “fix” it - so properly, in English, it’s a “bug fix” - but very often it’s shortened to “bugfix”.

Tryout my patch

sign out

sign in

ever present icon

There you’ll find a box housing the switch

except, as anticipated a little earlier, any custom services

user.timedout?(30.minutes.ago)

timeout

The exception, again, is the New York Times, where the one-word form is rare even on the blogs.

the spelling "Web site" (and the less questionable "web site") is an anachronism from the 1990s that is still in use by the NYT and some other conservative print media in the US while most others (including the online sections of the NYT!) today use "website".

Website is not a proper noun (as opposed to the Internet), hence not capitalized.

English tends to build new compound nouns by simply writing them as separate words with a blank. Once the compound is established (and the original parts somewhat "forgotten"), it's often written as one word or hyphenated. (Examples: shoelaces, aircraft...)

Web site / website seems to be somewhat in a transitional stage, being seen as an "entity" that web page hasn't reached yet. Depending on which dictionary you check you will find web site and website, but only web page, not webpage.

A website (also written as web site)

fallback

May I recommend the good old-fashioned non-confusing word capacity.

Different OS' have

The non-extension page, example.html

Anti-Features

Because they're more integrated and try to serialize an incomplete system (e.g. one with some kind of side effects: from browser/library/runtime versions to environment to database/API changes), they will tend to have high false-negatives (failing test for which the production code is actually fine and the test just needs to be changed). False negatives quickly erode the team's trust in a test to actually find bugs and instead come to be seen as a chore on a checklist they need to satisfy before they can move on to the next thing.