# XLSForm $ notation and select questions

**URL:** <https://forum.getodk.org/t/xlsform-notation-and-select-questions/10130>\
**Category:** Support\
**Tags:** odk-xlsform, form-design, odk-collect\
**Created:** [October 6, 2017, 3:07pm UTC](https://forum.getodk.org/t/xlsform-notation-and-select-questions/10130 "2017-10-06T15:07:31Z")\
**Posts on this page:** 1\
**Showing post:** 6

<div class="post-metadata">

**Author:** ![LN](https://getodk.b-cdn.net/user_avatar/forum.getodk.org/ln/32/1605_2.png) [@LN](https://forum.getodk.org/u/LN)\
**Post date:** [October 7, 2017, 5:23pm UTC](https://forum.getodk.org/t/xlsform-notation-and-select-questions/10130/6 "2017-10-07T17:23:08Z")

</div>

Unfortunately, I am pretty sure the answer is that this is not possible. You might also be tempted to use `pulldata` but the external itemset and `pulldata` implementations are completely unrelated and incompatible. This is related to our discussion about [interacting with data outside the form](https://forum.getodk.org/t/odk-1-0-developer-calls/7094/15) from the developer call a couple months back and illustrates why a single approach to external data that is complete will provide value. @dcbriccetti and @michal_dudzinski have been working on this (thanks!).

How many records are you loading, on what devices and how slow is it to use non-external `select_one`s? One step towards unifying the external data implementations is to improve performance internal `select_one` performance. If you can provide a sense of the size of your dataset (how many states, how many countries, how many cities) and the performance of the devices you work with, that will provide more information for that work. You can also follow it at [https://github.com/opendatakit/javarosa/issues/170](https://github.com/opendatakit/javarosa/issues/170)

---

_[View the full topic](https://forum.getodk.org/t/xlsform-notation-and-select-questions/10130)._
