Add correct link to eye camera position

core/best-practices/index.md

@@ -70,7 +70,7 @@
 **A.** False negatives - Actual fixations are not being detected, e.g. due to maximum dispersion too 
 low; minimum duration too high  
 **B.** False positives - Multiple actual fixations including their connecting eye movement
 (e.g. saccades) are being detected as a single fixation, e.g. due to maximum dispersion too high  
 **C.** Too many consecutive true positives that could be grouped to a single fixation, e.g. due to 
 maximum duration too low, or maximum dispersion too low
 
@@ -80,7 +80,7 @@
 classified fixations.
 
 1. A seated person fixates a static target. Their head is mostly stationary. Only small compensatory 
 eye rotations are needed to keep the target stabilised on the fovea. Fixations are long.
 
 In this task, setting maximum duration too low could lead to type C errors. If a fixation lasts 400ms, 
 filtering with a maximum duration of 200ms will classify two fixations instead of one. Ensure the maximum 
@@ -94,7 +94,7 @@
 (type B error). Ensure the minimum duration is appropriate for the viewing task.
 
 3. A dynamic environment where a walking person gazes at a sign. Their head is moving considerably when 
 compared to the previous tasks. Larger compensatory eye rotations are needed to stabilise the region of 
 interest on the fovea due to cyclic movements of the head and forward progression. The recorded gaze points 
 are dispersed over a greater area (this is an inherent issue with detecting fixations in scene camera space 
 instead of the real world). 
@@ -122,7 +122,7 @@
 ### Pupil Detection and Blinks
 It is worth noting that poor pupil detection in general can lead to false negatives. In such instances, adjusting the 
 thresholds can make it easier to detect blinks, but also increases the chance of false positives. It is worth taking the 
-time to ensure an optimal setup with regards to [eye camera positioning](/#_3-check-pupil-detection) and 
+time to ensure an optimal setup with regards to [eye camera positioning](/getting-started/#_3-check-pupil-detection) and 
 [2d detector settings](/software/pupil-capture/#fine-tuning-pupil-detection) so that the pupils are well-detected
 when the eyes are open.
 
@@ -147,10 +147,10 @@
 ### 1. Lab Streaming Layer (LSL)
 We maintain a [Plugin for LSL](https://github.com/labstreaminglayer/App-PupilLabs/tree/master/pupil_capture) that 
 publishes gaze data in real-time using the [LSL framework](https://labstreaminglayer.readthedocs.io/info/intro.html). 
 This enables unified and time-synchronised collection of measurements with other 
 [LSL supported devices](https://labstreaminglayer.readthedocs.io/info/supported_devices.html).
 
 The plugin synchronizes Capture's own clock, Pupil Time, with the pylsl.local_clock(). This allows the recording of 
 native Capture timestamps and removes the necessity of manually synchronizing timestamps after the effect, which makes
 for an accurate and easy to implement solution.
 
@@ -169,8 +169,8 @@
 [This script](https://github.com/pupil-labs/pupil-helpers/blob/master/python/remote_annotations.py) demonstrates how you 
 can send remote annotations over the network.
 
 ## Pupillometry
 [Pupillometry](https://doi.org/10.1002/wcs.1323) is the study of temporal changes in pupil diameter in response to 
 external light stimuli and/or cognitive processing. Pupil Core reports pupil diameter in mm provided by the 
 [pye3d](/developer/pye3d/#pye3d-pupil-detection) model: `diameter_3d`, and in pixels as observed in the eye 
 videos: `diameter`. 
@@ -210,7 +210,7 @@
 One solution to prevent 'Model Adjustment Error' is to freeze the model, which is available as a toggle in the eye windows. 
 You should only freeze the model when it is well-fitting as per the description above. Pupil size errors due to 
 'Actual Headset Slippage' can only be avoided by eliminating headset slippage, which should be a key aim when performing
 pupillometry in general, and especially when freezing the model.
 
 :::tip 
 You can also freeze the model when running