-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Link Dumps #60
Comments
https://facebook.github.io/react/blog/2015/05/01/graphql-introduction.html "Because of multiple round-trips and over-fetching, applications built in the REST style inevitably end up building ad hoc endpoints that are superficially in the REST style. These actually couple the data to a particular view which explicitly violates one of REST's major goals. |
Alan Kay on messaging: Hacker News discussion on Diaspora's 0.5 release: |
Hacker news on "the failure of agile" |
interested in your take on this one @ahdinosaur @joshuavial http://martinfowler.com/bliki/MicroservicePremium.html |
http://highscalability.com/blog/2014/4/8/microservices-not-a-free-lunch.html |
@simontegg i agree that microservices have a cost, which is why i started Holodex as a monolith and would recommend doing so for any project in the 'make it work' phase of development. the problem is that the longer a project remains a monolith (especially if it increases scope to include any dependency services), the harder it is to split out later. an open question that was brought up in the articles is 'what's the right size of a microservice?'. should Holodex be one service, many services? and that's not even mentioning the question of Holodex's scope. the approach i'm taking by intuition: clearly define the scope of the project, start building a monolith to solve this scope, then as things start to work split things out into re-usable modules and eventually de-coupled services, and if your scope ever requires another scope (Holodex will soon require authentication (who you are) which i've defined as out of scope for Holodex) please take the extra effort now to add it in a somewhat de-couple'd way as otherwise it's gonna be much more expensive to de-couple later, plus it's best for the ecosystem as a whole to have more re-usable services for common app dependencies. that's a bunch of rambling, does that help at all @simontegg? |
in the broader app ecosystem perspective, i'm okay if an app is grossly monolithic the first time some problem is solved (e.g. Loomio which is (should be?) scoped to discussions and decisions but also includes authentication, user profiles, groups (admins, memberships, invitations, profiles, child groups), notifications, etc.) but when it's the second time around (e.g. Cobudget which is (should be?) scoped to group crowdfunding but also includes authentication, groups (admins, memberships, invitations)) i find it harder to swallow, as i reckon that's when we should be putting in the extra effort to make the solution re-usable and de-coupled so that we solve the problem for good, not just until we want to build another app. |
|
|
http://makerbook.net : hand-picked directory of the best free resources for creatives. |
Oh, yes |
|
from Ants, ‘the future of work visualisation'. note the pricing is $13/user. |
via @rdbartlett
|
Yeah I keep re-learning over and over: information scarcity is the justification for command and control hierarchies >> information abundance creates opportunity for nonhierarchical organisation >> but it only works if you actually do the information abundance bit!! |
That (the gratipay conversation about maybe joining Enspiral) is really exciting. Please keep the valueflows gang informed? It's all part of the same convergence. Maybe. |
Wow. I assume that was a labor of love...but what a lot of both! |
https://medium.com/the-ready/the-org-chart-is-dead-e1d76eca9ce0#.tp3e1re6n
|
Looks like holodex to me. The author has apparently not caught on yet. |
^ agreed... and I find it somewhere between amusing and horrifying that they're openly trying to get funded by describing it as a billion dollar idea. |
The trouble with the expedient path and building for ourselves https://medium.com/swlh/the-surprisingly-simple-thing-slack-got-wrong-b16f489395e#.jge16ohq8 |
https://twitter.com/KentBeck/status/704452340397404160
|
^quite concise and clear |
Great link @simontegg. That definitely resonates with me on the last project I did. Oddly, the customer didn't seem like they were in a big rush, so I took the extra time (staying within the budget) to apply more refactorings ... a bunch of them. The result? A codebase I can honestly say I'm proud of. Based on the requirements, I knew exactly how the codebase will evolve and I designed it so those evolutions will be as easy and straightforward as possible. I attribute the quality of the codebase on my ability to go slow. And, it was a wonderful experience for me, and the customer is thrilled with the quality. Highly recommended. |
^ quite heartwarming :) thanks for sharing @locksmithdon !! |
Hammock Driven Development has been shared around these parts already, On Sat, 30 Apr 2016 at 14:20 Greg Cassel [email protected] wrote:
|
|
In one study (Barrow and Neville, 2001) People preferred traditional org charts and icicle plots to treemaps and tree ring charts for hierarchical data https://medium.com/@kennelliott/39-studies-about-human-perception-in-30-minutes-4728f9e31a73#.3m8kuq4f3 (scroll to halfway) cc @ahdinosaur PS starting to stew on holodex reboot |
cool! |
https://justinjackson.ca/saas/ SaaS is ripe for disruption |
Ask HN: What's the best 'non-frustrating' search/directory UI/UX you've seen? |
pulled some favorites from holodex/app#60
No description provided.
The text was updated successfully, but these errors were encountered: