Now that the Packs Builder has arrived, I just informed my family that I will not see them or any sunshine for the next 32 hours.
i so agree @Bill_French
i cannot wait to get my hands on the beta, then i too will be in ālock-downā while i explore the possibilities
max
Itās a game-changer for sure. It will open the floodgates of innovation, opportunity, and all while hyper-charging the aftermarket-place to rally around new use cases and more refined existing applications.
a long time ago in a galaxy far awayā¦ in the previous centuryā¦
i worked in a mobile software company (back in the days of the Blackberry, XDA & Psion).
i persuaded our engineers to add a small JScript interpreter to our data collection app.
it not only allowed our more advanced customers to INNOVATE like crazy,
but it allowed our ENGINEERS to create lots of new innovations quickly, prototyping them, beta testing them, getting them out there in a matter of days!
those new features that were successful were then re-engineered in the productās core.
but the rate of EVOLUTION for our product took off like a rocket!
so i predict that even CODANS will see a huge increase in INNOVATION and speed-to-market with these Power Packs
max
Looking forward to your ML in Coda!
Itās rolling out already? Iām still waiting for access
Yes. I think itās a trickle so far, but these Codans are on it.
Still waiting for access too but very excited for this. If anyone wants to collaborate on some packs, let me know.
Should be soon - and hereās a taste.
Damn! Please grant more access soon Coda
@Bill_French couple questions in the meantime:
- Have they said anything about which version of JS is supported (or perhaps which node version is running under the hood?)
- Can you import external dependencies?
Not that I am aware of.
I believe you can if you develop your packs with the CLI.
ryan (@Ryan_Redmond )
i understand the frustration behind your posts.
i too have applied for the beta program but have not yet been given access.
howeverā¦
i also recognize that coda has a beta program to run, and is being careful to roll out access in a controlled way. so i am prepared to remain patient, even though, just like yourself, i have client work that can benefit from this new capability.
so i accept your first post as stating your urgency for this beta program and your frustration at not gaining access yet, but i have flagged your posts as inappropriate because of the unacceptable level of aggression and disproportionate threat of (some kind of) retribution. not professional. not according to the guidelines for this community and not at all respectful to the hard working folks at coda and the rest of this very supportive community.
i would respectfully ask you to moderate your language whilst still remaining active in this community and expressing your opinions and suggestions, asking questions etc.
i hope my comments do not offend, but i hold to my position regardless.
respect
max
sorry i thought some of the sarcasm was apparentā¦apologies
Sarcasm factored in, I too have had periods of frustration with Coda and for that matter ā with all vendors who are reinventing what it means to build modern apps. The multi-dimensional paradox is that we want our platforms to work entirely different but remain unchanged and without any regression bugs. This is the fundamental requirement for ābetaā software, and from experience, itās not not easy.
In 1987 I had to sign off on a gold master to produce 53 million LapLink disks at a cost of $3 bucks each. Today, these manufacturing decisions are non-existent, but the responsibility and impact on businesses and users is vastly greater. In the 80ās, the compiled app on disk #1 did not have the capacity to break the app running on disk #2.
In terms of your own business needs I get that too. There are urgencies and a lot riding on any platform choice. However, like all beta products, there are known show-stoppers and likely pockets of code debt that Coda must consider. Furthermore, I can already see a number of changes that will probably pull the rug out from under some of the code Iām writing. This is usual and customary and Coda has been very transparent about these likely impactful changes. As good as the new Packs Studio is, itās not advised to create significant [immediate] dependencies your business can bank on. If you do, I can almost guarantee youāll be writing a different kind of disappointing message somewhere in this community.
@Ryan_Redmond, the hubris of me (someone with beta access) telling you (someone without beta access) to āchillā, does not escape me. But dude - yaā gottaā remain calm while the Codans work through the challenges and gather data while a constrained number of testers stress the platform enough before they put all users at risk.
ps - Itās worth the wait.
Haha Iām sure it is. The anticipation of what I want to create is killing me. I feel like (hope) this is the one element that was missing from codaā¦I couldnāt really pin what felt slightly empty about the doc , but this was most certainly it. Once again, I didnāt mean to draw any real attention earlier. I was speaking unfiltered.Which, was not necessaryā¦now Iāll never get beta access haha
Indeed. I had similar angst through the week following the Block Party and I was mentally pacing back and forth contemplating the possibilities. And imagine those of us waiting for this since 2019 as evidenced by threads like this one. But with this energy, I was able to make some progress before my beta request was accepted.
For example - given this simple code snippet I was able to deduce a general understanding of Packs and design a statistical outlier identifier; a simple formula that when given a list of numbers, it classifies the rows as normal or abnormal. Using the free Google Apps Script framework I was able to quickly create a test harness that mimics the addFormula() method and fully develop the [future] packās behavior. Once I was in, I tested my GAS-created pack facsimile in a few minutes and it worked.
The vision of an integrated Javascript framework in Coda was the first thing that came to mind in 2019 when I created my first Coda document and used formulas. I knew there was an extensibility issue but with the lineage of the creators, I also knew what the future might look like ā and here it is today.
The current Packs Studio is not a pathway that magically eliminates all Coda constraints, but it is a pretty amazing start and you can begin to build custom formulas today - long before you have access. Itās just javascript in an SDK environment that is [somewhat] documented - you are free to build and shore up those javascript skills today.
While statistical outliers are relatively simple, they provide a glimpse into machine learning. Iām using this pack to look for unusual responses from a customer survey built with Coda. The cool thing about this is the Pack is literally learning as the data flows in from our 2500 customers. With each new response, the definition of āoutlierā is subtly refined.
This demonstrates that even the simplest javascript implementations are capable of tiptoeing into the realm of data science and AI.
Have a look through the full documentation too - itās relatively comprehensive for something so new, and like @Bill_French says, you can likely get some initial code going (just not in a way you can easily run/test yet haha)
Thank you, this is what I was waiting for! Iām not really able to provide feedback to the beta team right now so I havenāt requested access. But it will be great to look over the documentation to get the ideas flowing anyway!