I’ve been using Coda for 3 weeks now, and I really like it. While I have many comments to communicate, I have a major one; that is memory management.
I use Coda with Safari, and using Coda as a Web app applies pressure on my Mac’s memory. When I do other memory-intensive tasks, it becomes an issue and I have to restart Safari from time to time.
Just imagine how two Coda docs have this significance on memory more than any other app:
I also notice that every Coda doc I open is treated like a complete dedicated instance of the app in my memory, and this is natural when opening the same Web app in two tabs. What if I want to open more docs? This is limiting me.
I recommend having a native app that handles its own docs and takes care of every thing, not via a browser. And if you built your app architecture around Web code, you can use one of the native techs that supports Web code like Electron. Just think how Visual Studio Code is so effective despite all work it does. I don’t think that Coda is more significant than it.
So please give this a high priority because it’s really a deal breaker, and I already started moving my systems from Excel to Coda.
You’re correct in that each Coda tab is running the full code that’s needed to use the doc. This is what makes it possible to keep working on a Coda doc that’s open in a tab even if you lose your internet connection. You’re also correct that it does use a good bit of memory.
A native app is something we’ve talked about and it is a possibility. Coda is still a young company and we’re growing in capability very quickly. It’s more a matter of when this makes it to the top of the list and there are a few things that need to be in place before that can happen. But rest assured, this is something that’s discussed.
I’m going to move this to the Suggestion Box so it surfaces when we search for members interested in this idea.
Thank you for the detailed feedback and suggestion!
We have 3 great help articles on improving performance in Coda that might help to troubleshoot issues with docs that slow down over time as rows are added…
Notion did a good job in that. Your app may be different in nature, but similar to it. When I compare Notion to an app like Evernote, I prefer Evernote as a fully native app. But when I want to use Notion, I prefer the app over using it via a Web browser.
You are right that Coda is still in its early years, but it’s very promising and unique.
I’m having another major issue.
I’m migrating a sheet from Excel. It contains right now about 392 rows with many calculating columns. I’ve prepared the empty table with all formulas. As I copy and paste columns from Excel to Coda, it takes forever to calculate. I copy all the 392 rows column by column at once.
If it will take time to calculate at first because of the number of rows I copy and then be fast it’s OK, despite having that Excel is not that slow. But will it recalculate all of that every time?
I’ve checked the optimization links above and still checking, also still at the beginning of moving data.
But if you charge Pro users for more than 1000 rows, how working with 500 rows be problematic this way?
My formulas may be complex, but it takes a second in Excel to be executed, not minutes.
Now while I’m typing this reply, the doc crashed for the second time while calculating.
This doc is one of the most important docs for me to move from Excel.
If there are formula dependencies on the rows, Load in a row and calculate, load another row and calculate, and so on. With complex formulas, this can be time-consuming.
Since moving from Excel to Coda requires adding the formulas, I’ve copied all data over first, then added the formulas after. If you already have a doc setup, you can create a new doc with a new table while keeping the same column names. Copy your data over. Then copy your formulas over one by one. This might be a lesser weight on your browser while you’re working through this migration.
I’ve also had a situation where I just moved 100 rows at a time. Once all data was in Coda though, everything was much easier to work with.
We actually did have your doc show up in our queue due to calculations running pretty hard. The issue has to do with a previous row formula that returns multiple results because its ranking on a non unique value. This is leading to lots of arrays of a nesting depth greater than 30. So it’s not a single calculation per row, things are running almost to not end.
If you can post a sample doc here of what you need, with maybe 3 rows of sample data, we might be able to help you to find better ways of getting this to work in Coda. I wouldn’t try to bring everything from a spreadsheet over to Coda in a direct sense, I would step back and see what Coda has to offer above and beyond spreadsheets and then try to bring the data over in a way that plays to those strengths.
Spreadsheets run on a per cell level and Coda tables are more of a database setup. So cumulative row formulas tend to clash with the typical schema that Coda is most use for. Here’s a great blog article on how and why Coda tables are different:
I understand what you mean. In another two docs migrated from Excel, I reconsidered all of their design and ended up with simpler, cleaner, and more powerful result. In these two docs, I started with clean data, mainly used for goal setting and time tracking that are starting a brand new year.
However, this doc is for a personal financial system in which my cycle started at July, and I had to move a lot of data. I thought if I have the same similar deign I’ll move data more quickly with simple copy and paste. I also was restricted as I stopped my Office subscription and have to move very fast.
I’ll reconsider my solution having your feedback above, and will prepare a demo doc sometime later.
I’ve created a copy of the doc with sample data. Somehow, formulas have been disabled in my doc, I don’t know why.
I log expenses and transactions and update multiple account values entry-by-entry. This is like banks statements but for more than one account. I used to do that with Excel for 5 years. I used a column to get the values from the row before referencing a unique index composed for date and time. I think this may be causing calculation issues. However, getting previous row data is problematic in Coda. I searched the community for solutions and ended up with what in this doc.
Now I can’t return formulas to work again despite I removed all problematic columns. Support chat is not available today. I can’t move my work forward. This is vert disappointing.
Thanks sharing the doc.
Unfortunately the formulas have been causing the system to block and after a quick look I come to the conclusion that the design of the schema should be amended as bases work some different from Excel / Gdocs.
Basically you should create one table with all income, a table with all expenses [transactions] and a balance table with a starting amount. By creating views you will be able to get specific for each account the relevant information.
Accordingly as you already created, you can for each income/expense assign the account it belongs to and in the balance it will become visible for each account type the saldo.
Next step is to connect the budget to your expenses and so on.
Maybe this post gives you some inspiration, with thanks to: @Saul_Garcia
@Mostafa_AlBelliehy, when calculations are disabled like that, there is usually a circular reference that is continuously calculating. If the system detects this, it will shut off calculations for your doc. Please do write into support and include the link to the doc that has the issue. Many times this takes an engineer to fix the issue.
I mentioned before that Coda uses Tables and not Spreadsheets. A big difference is the fact that formulas are column based and not cell based. Because of this, I would try to stay away from any “cumulative” formulas that rely on the previous row. I know a typical balance sheet looks this way, but I think there are better ways to use the data in Coda.
Hopefully to help with some of the memory overhead, I use Brave Browser and it seems to have much lower memory usage than what you’re experiencing, although it may be due to differences in the quantity of data needed for our documents.
I believe Brave has the lowest memory usage of most other browsers (although I don’t know how it stacks up against Safari).
