text stringlengths 0 1.36k |
|---|
• Introducing Sheetsee library for visualizing data from Google Spreadsheets |
• Use cases for Sheetsee, such as static websites and internationalization |
• Using spreadsheets as a settings page to generate websites |
• Utilizing Google Sheets features, such as GPS coordinates in addresses |
• Connection of Sheetsee to Tabletop.js for data retrieval and JSON generation |
• Handling scalability and potential server issues with Sheetsee |
• Integration with Glitch.com for easy server setup and backup |
• Possibility of syncing data locally using Pouch or service workers |
• Error handling in Tabletop, particularly dealing with failed Google Spreadsheet connections |
• JavaScript tooling often involves complex compile chains |
• Alex Sexton reminisced about building his first website at age 10 using members.aol.com and encountered difficulties with database concepts |
• He struggled to find information on retrieving data from a central repository without knowing the term "database" |
• Rachel White presented her pick of the week: Chaosbot, an experimental GitHub project that updates its own code through democratic voting |
• Mikeal Rogers is fascinated by Chaosbot's concept and has been following its development |
• Alex Sexton mentioned his favorite project on GitHub: Babili, a beta ES6 minifier for shipping modern JavaScript code to browsers |
• The speakers discuss their current projects and experiences with WebRTC and Node.js. |
• pkg, a new tool from Zeit, is introduced as a way to turn Node projects into single executable files. |
• A Medieval Fantasy City Generator tool is mentioned and shared in the live chat. |
• The speakers also mention their enthusiasm for using ES6 features in WebRTC experiments. |
**Mikeal Rogers:** Welcome to JS Party, where it's a party every week with JavaScript! I'm Mikeal Rogers... |
**Rachel White:** I'm Rachel White. |
**Alex Sexton:** And I'm Alex Sexton. |
**Mikeal Rogers:** And on the show today we also have Kat Marchán and Rebecca Turner from npm... Why don't you all say hi? |
**Kat Marchán:** Hi! |
**Rebecca Turner:** Hi! |
**Mikeal Rogers:** So let's just right into it. We really wanna talk about npm@5 today, we're really excited about it, so why don't you tell us a little bit of the back-story behind npm@5 and why this is such a big deal release? |
**Kat Marchán:** It's kind of a big deal... |
**Rebecca Turner:** It is! |
**Kat Marchán:** The story of npm@5 is that last October - or September...? When was it? |
**Rebecca Turner:** October. |
**Kat Marchán:** npm@4? So last October we released npm@4, and therefore we couldn't use the number four anymore, so we needed a bigger number to release, and we chose five as a valid increment for an integer. The further story is we've had a lot of breaking changes that we've been doing for a while, and that's usually... |
Sometime mid early last year we decided that we're just gonna bite the bullet and we're gonna do the cache rewrite. We'd been talking about it for five years, we have closed many issues with going like "This will be fixed by the mythical cache rewrite that we've been talking about for five years..." So we were like, "A... |
**Rebecca Turner:** Yeah, and then Kat started playing around with it in November... |
**Kat Marchán:** Yeah, it was November when I started looking at it... |
**Rebecca Turner:** ...and we first started to actually see results -- I think you tried it out in late January... |
**Kat Marchán:** Late January, early February, that was when I could actually do it. |
**Rebecca Turner:** ...and we were really surprised to find that it was unbelievably faster. The old cache was -- I still don't know how it was as slow as it was. |
**Kat Marchán:** It was slower than just fetching from the network sometimes, if it was faster at all. It was just not very fast. We still don't know why it was that slow... But it was! |
**Alex Sexton:** That's interesting... I'd be interested in that data. |
**Kat Marchán:** If you figure out what happened tell us, because I have no idea! \[laughter\] |
**Mikeal Rogers:** \[03:57\] So it's much faster... I don't think that's the only performance update though, right? There's a couple other things that you did to improve the performance around this time. |
**Rebecca Turner:** Yeah... I mean, that got us a 5x speed increase over older npms, and then it was like "Wow... That was a lot faster!" and now other improvements start to actually seem meaningful. What before would have been like "Well, that's like a 10% improvement in speed", but it was taking so long anyway that i... |
Probably the single biggest aspect of the speed in the new version is the new lockfile shrinkwrap support. Having those allows the install to be much faster. It's part of the reason we made a lockfile by default - it was just speed. |
**Kat Marchán:** Yeah, but npm@5 it was the speed, and I'd say the second one was usability improvements that we did. We were just like, "Alright, it's a major version. People care about usability - what can we do to make this tool easier to use?" And some insight here - there's a reason that we didn't save by default ... |
**Rebecca Turner:** Library authors were the majority of Node developers in 2012... Not so much now. |
**Kat Marchán:** Now we're pretty much \[unintelligible 00:05:42.27\] developers, right? |
**Alex Sexton:** In that vein, one of the things I was excited to see was the symlink stuff... Stripe, where I work, has a monorepo and we definitely have some jiu-jitsu around trying to move libraries that are in our thing into our dependencies, but have dependencies still work among our subdependencies, all in the sa... |
**Rebecca Turner:** Sure! So we've had this file specifier since npm@2, so you could npm install a local directory. It was added in at the very end of the npm@2 development cycle, and it wasn't super well integrated into the rest of the npm product. What went into the shrinkwrap was never fully specified and has varied... |
When we were working on designing npm@5, one of the things we wanted to do was to make the shrinkwrap situation, and thus the blockfiles generally made more sense and worked better. So one of the pieces of that was defining "What do file specifiers do?" and we were having a lot of problems figuring out how this should ... |
**Alex Sexton:** That's interesting. |
**Rebecca Turner:** The monorepo use case - there's still quite a bit of work to be done there as far as smoothing that out. I expect more on that the next coming six months. |
**Alex Sexton:** Cool. |
**Rebecca Turner:** We have a number of ideas; we think we'll make that better. Incidentally, internally, npm's own website has moved towards a monorepo, so we're getting to dogfood that pretty strongly now. |
**Alex Sexton:** \[07:58\] That's encouraging. |
**Mikeal Rogers:** Yeah, I have to say my favorite feature actually probably is the default save stuff. I think the first bug in most of my packages that I've received from other people is "You forgot to actually add this dependency" because I use save for, --save for half of the things that I installed, and then not o... |
**Kat Marchán:** Me too... I think we've been pushing to try and do this for a while... It's like, "How do we push this in?" This is a serious change in people's npm workflow, right? This is not a change that we could do lightly, as small as it seems... |
**Rebecca Turner:** It basically became non-optional when we decided to do the package locks. |
**Mikeal Rogers:** Yeah, let's get into that. That's a really, really big change in terms of default behavior, right? |
**Kat Marchán:** Yeah, so we're moving to optimize the default path for users as much as possible. We're also trying to cut back on how much configuration you can do and make things like configuration more binary seeming. So instead of saying "Here's your cache min, here's you cache max." Now you have three options --o... |
**Rebecca Turner:** The --cache-min 0. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.